TW201738758A - Instructions and logic to provide base register swap status verification functionality - Google Patents

Abstract

Instructions and logic provide base register swap status verification functionality. Embodiments include a processor having a first model specific register (MSR) to store a first base address corresponding to a segment for a first execution context and a second MSR to store a second base address corresponding to a segment for a second context. A third register stores a base register swap status field corresponding to the segment of the first and second contexts. A decode unit decodes a swap instruction and execution logic executes an exchange of the first MSR value and the second MSR value responsive to the swap instruction. The execution logic determines if said exchange of the first MSR value and the second MSR value completed successfully, and changes a value of the base register swap status field responsive to a determination that said exchange completed successfully.

Description

Instructions and logic for providing substrate register exchange state verification functions

The present disclosure pertains to the field of processing logic, microprocessors, and associated instruction set structures that perform logical, mathematical, or other functional operations when executed by a processor or other processing logic. In particular, the present disclosure is directed to instructions and logic for providing a substrate register exchange state verification function.

Background of the invention

Modern processors may provide a scratchpad to point to thread-specific data in one or more scratchpad files and/or module specific registers (MSRs) that sometimes include a scratchpad or multiple scratchpads.

The historic x86 architecture, for example, has eight general purpose registers (GPRs), six sector registers, a flag register, and an instruction indicator. Applications on most modern operating systems (similar to FreeBSD, Linux, or Microsoft Windows) use a memory module that points almost all of the extent registers to the same location (and uses paging to replace the area) Segment), effectively disabling the use of the segment registers. In general, the use of both segment registers FS and GS is an exception to this rule, which is used to point to thread-specific material. These can be used in 32-bit mode or in 64-bit mode. Some "system stylization" features of the x86 architecture are not used in 64-bit operating systems on some processors, or in so-called "long mode" (64-bit and compatible modes) - including The segment address (although the FS and GS segments are maintained in a residual form for use as an additional base indicator for the operating system architecture).

When FS and GS segment overrides are used in 64-bit mode, for example, their respective base addresses are used in linear address calculations: (FS or GS).Base (base) + index + offset. For the two-segment registers, FS and GS, the base address is set via two MSRs: FS.Base (C000_0100h) and GS.Base (C000_0101h). For example, in the long mode, one of the instructions called SwapGS can be used to exchange the contents of GS.Base with the contents of another MSR, KernelGSBase (C000_0102h). This instruction is used to reserve core information for use by one of the logical processor cores to bypass the context switch.

However, for some interrupts and/or exceptions, it is possible to reach between the core entry and the SwapGS instruction or between the SwapGS instruction and a core exit. Therefore, the exception handler may need to infer whether, at runtime, a SwapGS instruction needs to be executed -- sometimes, for example, to secure the system software. For example, some operating systems rely on ad hoc methods, such as complex and vulnerable "magic address check", or set the GS base register to a negative value in the core and enable the GS base register to be set. User space instructions are disabled. Some of these issues may add complications to the design of the exception handler, or require additional time-consuming checks and non-essential user restrictions.

As of today, possible solutions to this additional exception handler design complication, performance limitations issues, and system software security related issues have not been fully explored.

Summary of invention

According to an embodiment of the present invention, a processor is specifically provided, including: a first module specific register (MSR) for storing a corresponding one for a segment of a first execution context a first base address bar; a second MSR for storing a second base address field corresponding to the segment for a second execution context; a third temporary register for storing the corresponding a base register exchange status column to the segments of the first and second execution threads; a decoding unit for decoding a first exchange instruction; and an execution unit for performing the following actions: Performing an exchange of the value of the first MSR and the value of the second MSR on the decoded first exchange instruction, determining whether the exchange of the value of the first MSR and the value of the second MSR is successfully performed The completion, and the exchange of the value of the first MSR and the value of the second MSR are successfully determined by one of the completions to change one of the values of the base register exchange status column.

100‧‧‧ system

102‧‧‧Processor

104‧‧‧Cache memory

106‧‧‧Scratch file

108‧‧‧Execution unit

109‧‧‧Package Instruction Set

110‧‧‧Processor bus

112‧‧‧Graphics controller

114‧‧‧Accelerated graphics埠interconnect

116‧‧‧Memory Controller Hub

118‧‧‧ memory interface

120‧‧‧ memory

122‧‧‧System I/O

124‧‧‧Data storage

126‧‧‧Wireless transceiver

128‧‧‧Flash BIOS

130‧‧‧I/O Controller Hub

134‧‧‧Network Controller

140‧‧‧ computer system

141‧‧ ‧ busbar

142‧‧‧Execution unit

143‧‧‧Package Instruction Set

144‧‧‧Decoder

145‧‧‧Scratch file

146‧‧‧Synchronous Dynamic Random Access Memory (SDRAM) Control

147‧‧‧Static Random Access Memory (SRAM) Control

148‧‧‧ burst flash memory interface

149‧‧‧ PC Memory Card International Association / Compact Flash Card Control

150‧‧‧LCD display control

151‧‧‧Direct memory access controller

152‧‧‧ alternating bus main interface

153‧‧‧I/O busbar

154‧‧‧I/O Bridge

155‧‧‧Common asynchronous receiver/transmitter

156‧‧‧Common Sequence Bus (USB)

157‧‧‧Bluetooth Wireless UART

158‧‧‧I/O expansion interface

159‧‧‧ Processing core

160‧‧‧Data Processing System

161‧‧‧SIMD auxiliary processor

162‧‧‧Execution unit

163‧‧‧Instruction Set

164‧‧‧Scratch file

165‧‧‧Decoder

165B‧‧‧Decoder

166‧‧‧Main processor

167‧‧‧Cache memory

168‧‧‧Input/Output System

169‧‧‧Wireless interface

170‧‧‧ Processing core

171‧‧‧Auxiliary processor bus

200‧‧‧ processor

201‧‧‧Sequential front end points

202‧‧‧Quick Scheduler

203‧‧‧ Out-of-order execution engine

204‧‧‧Slow/general floating point scheduler

206‧‧‧Simple floating point scheduler

208, 210‧‧‧Scratch file

211‧‧‧Executive block

212‧‧‧ Address Generation Unit (AGU)

214‧‧‧ Address Generation Unit (AGU)

216‧‧‧fast ALU

218‧‧‧fast ALU

220‧‧‧Slow ALU

222‧‧‧Floating ALU

224‧‧‧Floating point mobile unit

226‧‧‧Instruction prefetcher

228‧‧‧ instruction decoder

230‧‧‧ Tracking cache

232‧‧‧Microcode ROM

234‧‧‧Micromanipulation queue

310‧‧‧Encapsulated Bytes

320‧‧‧Package blocks

330‧‧‧Package double word (dword)

341‧‧‧ Half-package

342‧‧‧ single package

343‧‧‧Double package

344‧‧‧Unsigned packed byte representation

345‧‧‧Signed Encapsulated Bytes Representation

346‧‧‧Unsigned packaged block representation

347‧‧‧Signed package block representation

348‧‧‧Unsigned packaged double-word representation

349‧‧‧Signed Encapsulated Double Word Representation

360‧‧‧Opcode format

361, 362‧‧‧ coding bar

363‧‧‧MOD column

364, 365‧‧‧ source operand identifier

366‧‧‧ Objective operand identifier

370‧‧‧Opcode format

371, 372‧‧‧ coding column

373‧‧‧MOD column

374, 375‧‧‧ source operand identifier

376‧‧‧ Objective Operator Identifier

378‧‧‧ prefix first tuple

380‧‧‧Opcode format

381‧‧‧ conditional column

382‧‧‧Operator bar

383‧‧‧Information size column

384‧‧‧Saturation type bar

385‧‧‧Source operator identifier

386‧‧‧ Objective Operator Identifier

387-389‧‧‧Operator bar

390‧‧‧Source operator identifier

391‧‧‧VEX prefix first tuple

392‧‧‧Operator bar

393‧‧‧ scalar-index-base identifier

394‧‧‧Selective displacement identifier

395‧‧‧Selective Instant Bits

396‧‧‧EVEX prefix first tuple

397‧‧‧Opcode format

398‧‧‧Opcode format

400‧‧‧Processor pipeline

402‧‧‧Select steps

404‧‧‧ Length decoding step

406‧‧‧ decoding steps

408‧‧‧Distribution steps

410‧‧‧Renaming steps

412‧‧‧ scheduling steps

414‧‧‧ scratchpad read/memory read

416‧‧‧Steps for implementation

418‧‧‧Write/Memory Write Step

422‧‧‧Exceptional processing steps

424‧‧‧Determining steps

430‧‧‧ front-end point unit

432‧‧‧ branch prediction unit

434‧‧‧ instruction cache unit

436‧‧‧ instruction conversion back buffer

438‧‧‧Command capture unit

440‧‧‧Decoding unit

450‧‧‧Execution engine unit

452‧‧‧Rename/Distributor Unit

454‧‧‧Decommissioning unit

456‧‧‧ Scheduler unit

458‧‧‧ entity register file unit

460‧‧‧Execution aggregation

462‧‧‧Execution unit

464‧‧‧Memory access unit

470‧‧‧ memory unit

472‧‧‧data TLB unit

474‧‧‧Data cache unit

476‧‧‧ Position 2 (L2) cache unit

490‧‧‧ processor core

500‧‧‧ processor

502A‧‧‧ core

502N‧‧‧ core

504A-N‧‧‧ cache unit

506‧‧‧Shared cache unit

508‧‧‧Integrated Graphical Logic

510‧‧‧System Agent Unit

512‧‧‧ring base interconnect unit

514‧‧‧Integrated memory controller unit

516‧‧‧ Busbar Controller Unit

600‧‧‧ system

610, 615‧‧ ‧ processor

620‧‧‧Graphic Memory Controller Hub

640‧‧‧ memory

645‧‧‧ display

650‧‧‧I/O Controller Hub (ICH)

660‧‧‧External graphic device

670‧‧‧ peripheral devices

695‧‧‧Front side busbars (FSB)

700‧‧‧Multiprocessor system

714‧‧‧I/O device

716‧‧ ‧ busbar

718‧‧ ‧ bus bar bridge

720‧‧ ‧ busbar

722‧‧‧ keyboard and / or mouse

724‧‧‧Audio I/O

727‧‧‧Communication device

728‧‧‧storage unit

730‧‧‧Digital and information

732‧‧‧ memory

734‧‧‧ memory

738‧‧‧High performance graphics circuit

739‧‧‧High-performance graphical interface

750‧‧‧ Point-to-point interconnect interface

752, 754‧‧‧P-P interface

770‧‧‧ processor

772‧‧‧Integrated memory controller unit

776, 778‧‧‧P-P interface

780‧‧‧ processor

782‧‧‧ integrated memory controller unit

786, 788‧‧‧P-P interface

790‧‧‧ chipsets

794, 798‧‧‧P-P interface

796‧‧‧ interface

800‧‧‧ system

814‧‧‧I/O device

815‧‧‧Remaining I/O devices

832, 834‧‧‧ memory

870, 880‧‧ ‧ processor

872, 882‧‧‧ Integrated Memory and I/O Control Logic (CL)

890‧‧‧ chipsets

900‧‧‧Single wafer system

902‧‧‧Interconnect unit

910‧‧‧Application Processor

920‧‧‧Media Processor

924‧‧‧Image Processor

926‧‧‧Optical processor

928‧‧‧Video Processor

930‧‧‧Static Random Access Memory (SRAM) Unit

932‧‧‧Direct Memory Access (DMA) Unit

940‧‧‧Display unit

1000‧‧‧ processor

1005‧‧‧CPU

1010‧‧‧GPU

1015‧‧‧Image Processor

1020‧‧‧Video Processor

1025‧‧‧USB controller

1030‧‧‧UART controller

1035‧‧‧SPI/SDIO Controller

1040‧‧‧ display device

1045‧‧‧High Definition Multimedia Interface (HDMI) Controller

1050‧‧‧MIPI controller

1055‧‧‧Flash memory controller

1060‧‧‧Double data rate controller

1065‧‧‧Security Engine

1070‧‧‧I2S/I2C (integrated inter-wafer sound/integrated circuit) interface

1110‧‧‧ hardware or software modules

1120‧‧‧ Simulation software

1130‧‧‧Storage

1140‧‧‧ memory

1150‧‧‧Wired connection

1160‧‧‧Wireless connection

1165‧‧‧Manufacture

1205‧‧‧Program

1210‧‧‧ Analog Logic

1215‧‧‧ processor

1302‧‧‧Higher language

1304‧‧x86 compiler

1306‧‧ x86 binary digital

1308‧‧‧Alternative Instruction Set Compiler

1310‧‧‧Alternative instruction set binary digital

1312‧‧‧Instruction Converter

1314‧‧‧ does not have the x86 instruction set core processor

1316‧‧‧with x86 instruction set core processor

1400‧‧‧ system

1410‧‧‧ memory

1412‧‧‧Translator or converter

1414‧‧‧ processor

1416‧‧‧Scratch file

1418‧‧‧Vector/FP register

1420‧‧‧General register

1422‧‧‧ register

1424‧‧‧Control register

1426‧‧‧Segment register

1428‧‧‧Task Status Section

1430‧‧‧ page

1432‧‧‧Descriptor Table

1501‧‧‧Based register exchange status verification function processing flow

1518-1524‧‧‧Processing steps

1502‧‧‧Based register exchange status verification function processing flow

1510-1526‧‧‧Processing steps

1503‧‧‧Based register exchange status verification function processing flow

1526-1550‧‧‧Processing steps

1601‧‧‧ Processor Microstructure

1602‧‧‧Processor microstructure

1603‧‧‧displacement

1604‧‧‧Base address

1605‧‧‧ index

1606‧‧‧Valid address

1608‧‧‧ mask

1609‧‧‧Information

Section 1610‧‧‧

1615‧‧‧Base register exchange status

1620‧‧‧EFLAGS physical register

1621‧‧‧Instruction Indicator Physical Register

1622‧‧‧MRS entity register

1624‧‧‧Control (CTL) physical register

1625‧‧‧CR3

1626‧‧‧Sector entity register

1627‧‧‧FS

1628‧‧‧GS

1630‧‧‧GS substrate

1632‧‧‧ nuclear GS substrate

1634‧‧‧FS base

1635‧‧‧MSR

1637‧‧‧Base

1638‧‧‧Base

1640‧‧‧Decoding unit

1650‧‧‧Execution engine unit

1652‧‧‧Rename/Distributor Unit

1654‧‧‧Demeritment unit

1656‧‧‧scheduler unit

1664‧‧‧Memory access unit

1670‧‧‧ memory unit

1672‧‧‧TLB

1674‧‧‧L1 cache

1676‧‧‧L2 cache

1680‧‧‧Floating point (FP) physical register

1682‧‧‧Mask physical register

1684‧‧‧ Vector entity register

1686‧‧‧Integer entity register

1694‧‧‧ Address generation logic

1699‧‧‧Storage data buffer

The invention is illustrated by way of example and not limited by the drawings.

1A is a block diagram of an embodiment of a system for executing instructions to provide a base register exchange state verification function.

FIG. 1B is a block diagram of another embodiment of a system that executes instructions to provide a base register exchange state verification function.

1C is a block diagram of another embodiment of a system that executes instructions to provide a base register exchange state verification function.

2 is a block diagram of an embodiment of a processor that executes instructions to provide a substrate scratchpad exchange state verification function.

FIG. 3A illustrates a package data pattern in accordance with an embodiment.

FIG. 3B illustrates a package data pattern in accordance with an embodiment.

FIG. 3C illustrates a package data pattern in accordance with an embodiment.

3D illustrates one of the instruction encodings for providing a substrate register exchange state verification function in accordance with an embodiment.

FIG. 3E illustrates one of the instruction encodings for providing a substrate scratchpad exchange state verification function in accordance with another embodiment.

FIG. 3F illustrates one of the instruction encodings for providing a substrate register exchange state verification function in accordance with another embodiment.

FIG. 3G illustrates one of the instruction encodings for providing a substrate register exchange state verification function in accordance with another embodiment.

FIG. 3H illustrates one of the instruction encodings for providing a substrate register exchange state verification function in accordance with another embodiment.

4A illustrates elements of a processor embodiment of a microstructure that performs instructions to provide a substrate scratchpad exchange state verification function.

4B illustrates elements of another processor embodiment of a microstructure that performs instructions to provide a substrate scratchpad exchange state verification function.

5 is a block diagram of a processor embodiment of executing instructions to provide a base register exchange state verification function.

6 is a block diagram of an embodiment of a computer system that executes instructions for providing a base register exchange state verification function.

7 is a block diagram of another embodiment of a computer system that executes instructions for providing a base register exchange state verification function.

8 is a block diagram of another embodiment of a computer system that executes instructions for providing a base register exchange state verification function.

9 is a block diagram of an embodiment of a single wafer system that executes instructions for providing a substrate register exchange state verification function.

10 is a block diagram of a processor embodiment of executing instructions to provide a base register exchange state verification function.

11 is a block diagram of an embodiment of an IP core development system that provides a base register exchange status verification function.

Figure 12 illustrates an embodiment of a structural simulation system that provides a substrate register exchange status verification function.

Figure 13 illustrates an embodiment of a system for converting instructions that provide a base register exchange state verification function.

Figure 14 illustrates a different embodiment of a system for converting instructions that provide a base register exchange status verification function.

Figure 15A illustrates a flow diagram of an embodiment of a handler for providing a substrate scratchpad exchange status verification function.

Figure 15B illustrates a flow diagram of various embodiments of a process for providing a one of the base register exchange status verification functions.

Figure 15C illustrates a flow chart of another different embodiment of a process for providing a substrate register exchange status verification function.

Figure 16A illustrates elements of a processor microstructure embodiment that executes instructions to provide a substrate register exchange state verification function.

Figure 16B illustrates a more detailed element of a processor microstructure embodiment that executes instructions to provide a substrate scratchpad exchange state verification function.

Detailed description of the preferred embodiment

The following description discloses instructions and logic for providing a base register exchange state verification function. Some embodiments may include a processor including: a first module specific register (MSR) to store a first base address field corresponding to a segment for a first execution context and a The second MSR stores a second base address field corresponding to the segment for use in a second execution context. A third register stores a base register exchange status column corresponding to the segments of the first and second execution threads. The processor decoding unit decodes a sector exchange instruction and performs a logical response to the decoded sector exchange instruction to perform an exchange of the first MSR value and the second MSR value. If it is determined that the exchange of the first MSR value and the second MSR value is successfully completed, the execution logic responds to the determination that the exchange of the first MSR value and the second MSR value is successfully completed. And change the value of one of the base register exchange status bars.

Historically, it has not been possible to have an unlicensed handler write an arbitrary value to the FS base register or GS base register. Two new instructions, WrFSBase and WrGSBase, now allow an operating system to actuate the unlicensed handler to write to the FS base register and the GS base register, respectively. Therefore, for an exceptional handler, it is increasingly difficult to determine if a substrate register has been set to access a value of the core information. It will be appreciated that, as described herein, the substrate register exchange state verification can be used to provide an exception handler with the ability to infer at runtime, for example, whether a SwapGS instruction needs to be executed without resorting to An ad hoc method, such as a complex and vulnerable "magic address check", or setting the GS base register to a negative value in the core and for setting the FS base register and/or GS base temporary storage User space instructions are disabled, for example, WrFSBase and WrGSBase. It will also be appreciated that the base register swap state verification instructions can be used to avoid complication of exception handler designs, additional time consuming checks, and unnecessary user restrictions.

In the following description, numerous specific details, such as processing logic, processor types, microstructure conditions, events, actuating mechanisms, and the like, are referred to in order to provide a more complete understanding of the embodiments of the invention. However, it will be apparent to those skilled in the art that the present invention may be practiced without the specific details. Additionally, some of the conventional structures, circuits, and the like are not shown in detail to avoid obscuring the embodiments of the present invention.

Although the following embodiments are described with reference to a processor, other embodiments are applicable to other types of integrated circuits and logic devices. Similar techniques and principles of embodiments of the present invention can be applied to other types of circuits or half Conductor devices that can benefit from larger addressable memory, higher pipeline throughput, and improved safety system performance. The techniques of embodiments of the invention are applicable to any processor or machine that performs data processing. However, the present invention is not limited to a processor or machine that operates on 512-bit, 256-bit, 128-bit, 64-bit, 32-bit, or 16-bit data, and can be applied to data processing or Manage any processors and machines that are being executed. Moreover, the following description provides examples and the figures show various examples for illustrative purposes. However, the examples are not to be considered as limiting, as they are merely intended to provide an embodiment of the invention and are not intended to provide an exhaustive list of all possible embodiments of the embodiments of the invention.

Although the following examples illustrate instruction processing and allocation in the context of execution units and logic circuits, other embodiments of the present invention may be implemented by storing data and/or instructions on a machine readable, materialized medium, where When the data and/or instructions are made using the machine, the machine will be caused to perform functions consistent with at least one embodiment of the present invention. In one embodiment, the functions associated with embodiments of the present invention are implemented in machine-executable instructions. The instructions can be used to cause the general inventive or special purpose processor programmed by the instructions to perform the steps of the present invention. Embodiments of the invention may be provided as a computer program product or software, which may include a machine or computer readable medium having instructions stored thereon that may be used to program a computer (or other electronic device) One or more operations are performed in accordance with an embodiment of the present invention. Additionally, the steps of an embodiment of the invention may be performed by a particular hardware component that includes fixed function logic for performing the steps, or by any combination of a stylized computer component and a fixed function hardware component.

Instructions used to program logic to carry out embodiments of the present invention can be stored in a memory in the system, such as DRAM, cache, flash memory, or other storage. Still further, the instructions can be distributed via a network or via other computer readable media. Thus, the machine readable medium may include any mechanism for storing or transmitting information in a form readable by a machine (eg, a computer), but is not limited thereto, a floppy disk, a compact disc, or a compact disc. , read-only memory (CD-ROM), and ferromagnetic optical disc, read-only memory (ROM), random access memory (RAM), can eliminate programmable read-only memory (EPROM), electrical can be eliminated Program-controlled read-only memory (EEPROM), magnetic or optical card, flash memory, or used to transmit signals (eg, carrier waves, infrared signals, digital signals, etc.) via electrical, optical, audible, or other means One of the physical machines on the Internet can read the storage. Thus, computer-readable media includes any type of materialized machine readable medium suitable for storing or transmitting electronic instructions or information in a form readable by a machine (eg, a computer).

A design can go through various stages, from creation to simulation to manufacturing. Information indicating a design can represent the design in some way. First, as useful when simulating, the hardware can be represented using a hardware narrative language or another functional narrative language. Additionally, a circuit level module having logic and/or transistor gates can be generated during some stages of the design process. Further, most designs, at some stages, reach the data level of the actual placement of the various devices representing the hardware modules. In the case where the semiconductor manufacturing technology is used, the data indicating the hardware module can be used to produce In the mask of the biofilm circuit, information indicating the presence or absence of various features of different mask layers. In any design representation, the material may be stored in any form of machine readable media. A memory or a magnetic or optical storage, such as a magnetic disk, may be a machine readable medium whose storage is transmitted via modulation or otherwise optically or electrically generated in a different manner to transmit the information. Information. When an electrical carrier that indicates or carries a digital or design is transmitted, to some extent, the copying, buffering, or retransmission of the electrical signal is performed, and a new copy is made. Thus, a communication provider or network provider may, at least temporarily, store an object (e.g., information encoded as a carrier to implement the techniques of the embodiments of the present invention) on a physical machine readable medium.

In modern processors, a number of different execution units are used to process and execute a variety of digital and instructional. Not all instructions are generated the same, some are completed faster and others may take some clock cycles to complete. The faster the commanded production capacity, the better the overall performance of the processor. It would therefore be advantageous to execute as many instructions as possible as quickly as possible. However, there are certain instructions that have greater complexity and require more execution time and processor resources. For example, there are floating point instructions, load/store operations, data movement, and the like.

As more and more computer systems are used in Internet, text, and multimedia applications, additional processor support has often been introduced. In one embodiment, an instruction set may be associated with one or more computer structures including data patterns, instructions, scratchpad structures, addressing modes, memory structures, interrupt and exception handling, and external inputs and outputs ( I/O).

In one embodiment, an instruction set architecture (ISA) may be implemented using one or more microstructures including processor logic and circuitry used to implement one or more instruction sets. Thus, processors having different microstructures can share at least a portion of a common instruction set. For example the same version, the Intel Pentium 4 (Intel®Pentium4) processor, Intel Core (Intel®Core ^TM) processors from Advanced Micro Devices, and the processor means Company of Sunnyvale, California, is almost carry out the x86 instruction set (and has Add some extensions to newer versions), but with different internal designs. Similarly, processors designed by other processor development companies (eg, ARM Holdings, MIPS, or their licensors or adopters) can share at least a portion of a common instruction set, but can include different processing. Design. For example, the same scratchpad structure of the ISA can be implemented in different microstructures in different ways using new or conventional techniques, including a dedicated physical scratchpad, one or more renaming mechanisms using a scratchpad (eg, A physical register that is dynamically allocated by a register aliasing list (RAT), a reordering buffer (ROB), and a deduplication register file. In one embodiment, the scratchpad may contain one or more registers, scratchpad structures, scratchpad files, or other sets of registers that may or may not be addressable by a software programmer.

In an embodiment, an instruction may include one or more instruction formats. In one embodiment, an instruction format may indicate various fields (number of bits, bit positions, etc.) to indicate among other things, the operations to be performed and the operations on which the operations are to be performed. Some instruction formats can be further defined using instruction patterns (or sub-formats). For example, one gives The instruction pattern of the instruction format can be defined to have a different subset of the fields of the instruction format and/or be defined to have differently interpreted fields that are given. In one embodiment, an instruction is represented using an instruction format (and, if defined, one of the instruction patterns of the instruction format) and specifies or indicates an operation and the operation is to be performed thereon The operand of the operation.

Science, finance, automated vectorization general purpose, RMS (identification, mining, and synthesis), and visual and multimedia applications (eg, 2D/3D graphics, image processing, video compression/decompression, voice recognition algorithms, and audio processing) ) It may be necessary to perform the same operation on a large number of data items. In one embodiment, single instruction multiple data (SIMD) refers to a type of instruction that causes a processor to perform an operation on a plurality of data elements. SIMD techniques can be used in a processor that can logically partition a bit in a register into fixed or variable size data elements, each representing a respective value. For example, in one embodiment, a bit in a 64-bit scratchpad can be organized as a source operand containing one of four distinct 16-bit data elements, each representing a respective 16-bit element. value. This type of material can be referred to as a 'packaged' data type or a 'vector' data type, and the data elements of this data type are referred to as package data operands or vector operation elements. In one embodiment, a package data item or vector may be a sequence of encapsulated data elements stored in a single scratchpad, and a packaged data operand or a vector operation element may be a SIMD instruction (or 'package data') One of the source or destination operands of the instruction 'or a 'vector instruction'). In one embodiment, a SIMD instruction designation is to be made to two source vectors. A single vector operation on an operand to produce a destination vector operand (also referred to as a result vector operand) having the same or different number of data elements and the same or different size of the same or different data element order. .

SIMD technology, for example, has been comprising x86, Intel®Core ^TM processor MMX ^TM, streaming SIMD extensions (SSE), SSE2, SSE3, SSE4.1, one SSE4.2 instruction and the instruction set employed, e.g., Used by ARM processors with the ARMCortex® family of processors that include one of the vector floating point (VFP) and/or NEON instructions, and, for example, by Loongson, developed by the Computer Technology Association (ICT) of the China Institute of Science and Technology. Adopted by the family of MIPS processors, with one of the major improvements in enabling application performance (Core ^TM and MMX ^TM are registered trademarks or trademarks of Intel Corporation of Santa Clara, Calif.).

In one embodiment, the purpose and source register/data is a generic term that indicates the source and purpose of the corresponding data or operation. In some embodiments, they may be implemented by a scratchpad, memory, or other storage area having other names or functions than those shown. For example, in an embodiment, "DEST1" may be a temporary storage register or other storage area, and thus "SRC1" and "SRC2" may be a first and second source storage register or other storage area, And others. In other embodiments, two or more SRC and DEST storage areas may correspond to different data storage elements within the same storage area (eg, a SIMD register). In one embodiment, one of the source registers may also act as a destination buffer, for example, by being on the first and second source materials. One of the operation results is written back to one of the two source registers as a destination register.

1A is a block diagram of an example of a computer system formed by a processor including an execution unit that executes an instruction, in accordance with an embodiment of the present invention. For example, in an embodiment illustrated herein in accordance with the present invention, system 100 includes a component, such as processor 102, to employ an execution unit that includes logic to perform algorithms for processing data. System 100 is available from representatives based on Intel's Santa Clara, California's ^{PENTIUM ® III, PENTIUM ® 4,} Xeon TM, Itanium ®, XScale TM and / or StrongARM ^TM microprocessor processing system, although other systems ( PCs with other microprocessors, engineering workstations, set-top boxes, and the like can also be used. In a While other operating systems (eg, UNIX and Linux), embedded software, and / or graphics embodiments, the sample system 100 may perform a version of WINDOWS ^TM systems available from Microsoft Corporation of Redmond, Washington, The user interface can also be used. Therefore, the embodiments of the present invention are not limited to any specific combination of hardware circuits and software.

Embodiments are not limited to computer systems. Different embodiments of the invention may be used in other devices, such as handheld devices and embedded applications. Some examples of handheld devices include mobile phones, internet protocol devices, digital cameras, personal digital assistants (PDAs), and handheld PCs. The embedded application may include a microcontroller, a digital signal processor (DSP), a single chip system, a network computer (NetPC), a set-top box, a network hub, a wide area network (WAN) switch, or may be based on at least one Example one Any other system of one or more instructions.

1A is a block diagram of a computer system 100 formed by a processor 102 that includes one or more execution units 108 for performing an algorithm to perform at least one instruction, in accordance with an embodiment of the present invention. An embodiment may be illustrated in the context of a single processor desktop or server system, although different embodiments may be included in a multi-processor system. System 100 is an example of a 'hub' system architecture. The computer system 100 includes a processor 102 for processing data signals. The processor 102 can be, for example, a Complex Instruction Set Computer (CISC) microprocessor, a Reduced Instruction Set Computing (RISC) microprocessor, a Very Long Instruction Block (VLIW) microprocessor, and an instruction to execute an instruction. A combined processor, or any other processor device, such as a digital signal processor. The processor 102 is coupled to a processor bus 110 that can transmit data signals between the processor 102 and other components of the system 100. The elements of system 100 perform those functions that are familiar to those skilled in the art.

In one embodiment, the processor 102 includes a one-bit 1 (L1) internal cache memory 104. Depending on its structure, the processor 102 can have a single internal cache or a majority of internal caches. Additionally, in another embodiment, the cache memory may be external to the processor 102. Other embodiments may also include a combination of both internal and external caches, depending on the particular implementation and needs. The scratchpad file 106 can store different types of data in various scratchpads, including an integer register, a floating point register, a status register, and an instruction indication register.

An execution unit 108 that includes logic for integer and floating point operations, the same It also exists in the processor 102. The processor 102 also includes a microcode (ucode) ROM that stores microcode for certain macro instructions. For an embodiment, execution unit 108 includes logic to process package instruction set 109. The operations used by many multimedia applications can be used with the package material in the general purpose processor 102 by including the package instruction set 109 in the instruction set of the general purpose processor 102, and the associated circuitry that executes the instructions. get on. Thus, many multimedia applications can be accelerated and executed more efficiently by using the full width of the data bus of the processor used to perform the operations on the packaged data. This eliminates the need to transfer smaller unit data across the processor's data bus to perform one or more operations per data element.

Different embodiments of an execution unit 108 can also be used in microcontrollers, embedded processors, graphics devices, DSPs, and other types of logic circuits. System 100 includes a memory 120. The memory 120 can be a dynamic random access memory (DRAM) device, a static random access memory (SRAM) device, a flash memory device, or other memory device. Memory 120, which may store instructions and/or data represented by data signals that may be executed by processor 102.

A system logic wafer 116 is coupled to the processor bus 110 and memory 120. In the illustrated embodiment, the system logic chip 116 is a memory controller hub (MCH). The processor 102 can communicate to the MCH 116 via a processor bus 110. The MCH 116 provides a high bandwidth memory path 118 to the memory 120 for storage of instructions and data and for storage of graphics commands, data and texture structures. The MCH 116 It is used to guide data signals between the processor 102, the memory 120, and other components in the system 100, and is bridged between the processor bus 110, the memory 120, and the system I/O 122. signal. In some embodiments, the system logic chip 116 can provide graphics for coupling to a graphics controller 112. The MCH 116 is coupled to the memory 120 via a memory interface 118. Graphics card 112 is coupled to the MCH 116 via an accelerated graphics layer (AGP) interconnect 114.

System 100 uses a proprietary hub interface bus 122 to couple the MCH 116 to the I/O controller hub (ICH) 130. The ICH 130 provides direct connection to some I/O devices via a local area I/O bus. The local I/O bus is a fast I/O bus for connecting peripherals to the memory 120, the chipset, and the processor 102. Some examples are audio controllers, firmware hubs (flash BIOS) 128, wireless transceivers 126, data storage 124, conventional I/O controllers with user input and keyboard interfaces, and a series of expansion ports (eg, A universal series bus (USB)), and a network controller 134. The data storage device 124 can include a hard disk drive, a floppy disk drive, a CD-ROM device, a flash memory device, or other mass storage device.

For another embodiment of the system, instructions in accordance with one embodiment can be used in conjunction with a single wafer system. One embodiment of a single wafer system includes a processor and a memory. The memory for this system is a flash memory. The flash memory can be placed on the same crystal form as the processor and other system components. Additionally, other logic blocks, such as a memory controller or graphics controller, may also be placed on a single wafer system.

FIG. 1B illustrates a data processing system 140 that implements the principles of an embodiment of the present invention. Those skilled in the art will appreciate that the embodiments described herein can be utilized in various processing systems without departing from the scope of the embodiments of the invention.

Computer system 140 includes a processing core 159 that can execute at least one instruction in accordance with an embodiment. For an embodiment, processing core 159 represents a processing unit of any type of architecture, including but not limited to a CISC, a RISC, or a VLIW type of structure. Processing core 159 may also be adapted to be fabricated in one or more processing techniques and expressed in sufficient detail on a machine readable medium, but may be adapted to facilitate the manufacture.

Processing core 159 includes an execution unit 142, a set of scratchpad files 145, and a decoder 144. Processing core 159 also includes additional circuitry (not shown in the graphics) that is not required for an understanding of embodiments of the present invention. Execution unit 142 is used to execute instructions received by processing core 159. In addition to performing general processor instructions, execution unit 142 may also execute instructions in package instruction set 143 for operation on the package data format. The package instruction set 143 includes instructions for performing embodiments of the present invention, as well as other package instructions. Execution unit 142 is coupled to register file 145 by an internal bus. The scratchpad file 145 represents a storage area on the processing core 159 for storing information containing the data. As previously stated, it should be understood that the storage area used to store packaged data is not critical. Execution unit 142 is coupled to decoder 144. Decoder 144 is used for instructions to be received by processing core 159 The decoding becomes a control signal and/or a microcode entry point. In response to these control signals and/or microcode entry points, execution unit 142 performs the appropriate operations. In one embodiment, the decoder is used to translate the opcode of the instruction, which will indicate which operations should be performed on the corresponding material indicated within the instruction.

Processing core 159 is coupled to bus 141 for communication with various other system devices, which may include, but is not limited to, for example, synchronous dynamic random access memory (SDRAM) control 146, static random access memory ( SRAM) control 147, burst flash memory interface 148, PC Memory Card International Association (PCMCIA) / Compact Flash (CF) card control 149, liquid crystal display (LCD) control 150, direct memory access (DMA) The controller 151, and the alternate bus master interface 152. In one embodiment, data processing system 140 may also include an I/O bridge 154 for communicating with various I/O devices via an I/O bus 153. Such I/O devices may include, but are not limited to, for example, a Universal Non-Synchronous Receiver/Transmitter (UART) 155, a Universal Serial Bus (USB) 156, a Bluetooth Wireless UART 157, and an I/O Expansion Interface 158.

One embodiment of data processing system 140 provides for mobile, network, and/or wireless communication, and processing core 159 can perform SIMD operations that include a string comparison operation. The processing core 159 can be programmed by various audio, video, video, and communication algorithms including discrete conversions, for example, a Walsh-Hadamard conversion, a fast Fourier transform (FFT), and a Discrete cosine transform (DCT), and their respective inverse transforms; compression/decompression techniques, for example, color space conversion, view The coded motion estimation or video decoding motion compensation; and the modulation/demodulation function, for example, Pulse Code Modulation (PCM).

1C illustrates another different embodiment of a data processing system that can execute instructions to provide a substrate scratchpad exchange status verification function. According to a different embodiment, data processing system 160 can include a main processor 166, a SIMD auxiliary processor 161, a cache memory 167, and an input/output system 168. The input/output system 168 can be selectively coupled to a wireless interface 169. The SIMD auxiliary processor 161 is operable to include instructions in accordance with an embodiment. Processing core 170 may be adapted to be fabricated in one or more processing techniques and represented on a machine readable medium in sufficient detail, but may be adapted to facilitate data processing including all or a portion of processing core 170. Fabrication of system 160.

For an embodiment, the SIMD auxiliary processor 161 includes an execution unit 162 and a register file 164. One embodiment of main processor 166 includes a decoder 165 to identify instructions that include an instruction set 163 for executing instructions by execution unit 162 in accordance with an embodiment. For different embodiments, SIMD auxiliary processor 161 also includes at least a portion of decoder 165B to decode the instructions of instruction set 163. Processing core 170 also includes additional circuitry (not shown in the figures) which is not required for an understanding of embodiments of the invention.

When operating, main processor 166 executes a stream of data processing instructions that control the general type of data processing operations that include interaction with cache memory 167 and input/output system 168. Embedded within the data processing instruction stream is a SIMD auxiliary processor instruction. Main processor 166 Decoder 165 acknowledges these SIMD auxiliary processor instructions, as would be performed by a SIMD-assisted processor 161. Thus, main processor 166 issues these SIMD auxiliary processor instructions (or control signals representative of SIMD auxiliary processor instructions) from their auxiliary processor bus 171 received by any of the SIMD auxiliary processors. In this case, the SIMD assist processor 161 will accept and execute any of the received SIMD assist processor instructions for it.

The data may be received via wireless interface 169 for processing by the SIMD auxiliary processor instructions. For an example, voice communication may be received in the form of a digital signal that may be processed using SIMD assisted processor instructions to regenerate a digital audio sample representation of the voice communication. For another example, the compressed audio and/or video may be received in the form of a digital bit stream that may be processed using SIMD assisted processor instructions to regenerate the digital audio samples and/or the moving video frame. For one embodiment of the processing core 170, the main processor 166, and a SIMD auxiliary processor 161 are integrated into a single processing core 170 including an execution unit 162, a register file 164, and a decoder 165 to An instruction is identified that includes an instruction set 163 in accordance with an instruction of an embodiment.

2 is a block diagram of a microstructure of a processor 200 including logic circuitry for performing instructions in accordance with an embodiment of the present invention. In some embodiments, instructions in accordance with an embodiment may be implemented to have data elements of a size, such as a byte, a block, a double block, a quad, or the like, and a data pattern, such as single and double precision. Sexual integers and operations on floating point data types. In an embodiment, the sequential front end point 201 is part of the processor 200. A copy that fetches the instructions to be executed and prepares them for later use in the processor pipeline. The front end point 201 can contain a number of units. In one embodiment, instruction prefetcher 226 fetches instructions from memory and feeds them to an instruction decoder 228, which then decodes or translates them. For example, in one embodiment, the decoder decodes a received instruction into one or more operations executable by a machine called a "microinstruction" or "micro-operation" (also referred to as micro-op or uop). . In other embodiments, the decoder parses the instructions into an opcode and corresponding material, and a control bar used by the microstructure to perform operations in accordance with an embodiment. In one embodiment, the trace cache 230 employs decoded micro-ops and combines them into a program ordered sequence or trace for execution in the micro-ops array 234. When the trace cache 230 encounters a complex instruction, the microcode ROM 232 provides the micro-ops needed to complete the operation.

Some instructions are converted into a single micro-operation, while others require many micro-ops to perform all operations. In one embodiment, if more than four micro-ops are required to complete an instruction, decoder 228 accesses micro-coded ROM 232 to process the instruction. For an embodiment, an instruction can be decoded into a small number of micro-ops for processing at instruction decoder 228. In another embodiment, an instruction can be stored within the microcode ROM 232 if some micro-operation is required to achieve the operation. Tracking cache 230 refers to an entry point programmable logic array (PLA) to determine one of the correct microinstruction indicators for reading a microcode sequence from microcode ROM 232 to complete one or more instructions in accordance with an embodiment. After the microcode ROM 232 ends the sorting for the micro-ops of an instruction, the endpoint 201 is restarted before the machine Self-tracking cache 230 takes micro-operations.

The out-of-order execution engine 203 is where the instructions are prepared for execution. The out-of-sequence execution logic has buffers to smooth out and reorder the instructions so that they can optimize performance as they go down to the pipeline and get scheduled for execution. The allocator logic allocates each micro-op in order to perform the required machine buffers and resources. The scratchpad rename logic renames the logical scratchpad to an item in a scratchpad file. The distributor also assigns an item to each of the two micro-operation queues (one for memory operation and one for non-memory operation), before the instruction scheduler: memory A volume scheduler, a fast scheduler 202, a slow/general floating point scheduler 204, and a simple floating point scheduler 206. The micro-ops schedulers 202, 204, 206 determine when a micro-op is ready to execute based on the readiness of their slave input register operand sources and the effectiveness of the execution resources of the micro-operations that need to complete their operations . The fast scheduler 202 of an embodiment can schedule every half cycle of the main clock cycle, while other schedulers can only schedule one cycle per master processor clock cycle. The scheduler arbitrates the micro-operations that are scheduled for execution.

The scratchpad files 208, 210 are located between the schedulers 202, 204, 206, and the execution units 212, 214, 216, 218, 220, 222, 224 in the execution block 211. It is a separate register file 208, 210 for integer and floating point operations, respectively. Each of the scratchpad files 208, 210 of an embodiment also includes a bypass network that bypasses or transmits the results of the slave microprocessor file that has just been completed but not yet written to the scratchpad file. The integer register file 208 and the floating point register file 210 are also Can communicate with other people. For an embodiment, the integer register file 208 is separated into two separate scratchpad files, a scratchpad file is used for low-order 32-bit data and a second scratchpad file is used for High-order 32-bit data. The floating point register file 210 of an embodiment has a 128 bit wide item because floating point instructions typically have operands from a width of 64 to 128 bits.

Execution block 211 contains execution units 212, 214, 216, 218, 220, 222, 224, where the instructions are actually executed. This portion contains the scratchpad files 208, 210, which store the integers required for microinstruction execution and the floating point data operand values. The processor 200 of an embodiment includes some execution units: an address generation unit (AGU) 212, an AGU 214, a fast ALU 216, a fast ALU 218, a slow ALU 220, a floating point ALU 222, and a floating point mobile unit 224. For an embodiment, floating point execution blocks 222, 224 perform floating point, MMX, SIMD, and SSE, or other operations. The floating point ALU 222 of an embodiment includes a 64 bit x64 bit floating point divider to perform the division, the square root, and the remainder of the micro operations. For embodiments of the present invention, instructions relating to a floating point value may be processed by floating point hardware. In one embodiment, the ALU operation proceeds to the fast ALU execution unit 216, 218. The fast ALUs 216, 218 of an embodiment can perform fast operations by one of the half clock cycles of an effective latency. For an embodiment, most of the complex integer operations go to the slow ALU 220, since the slow ALU 220 contains integer execution hardware for long latency type operations, such as a multiplier, shifter, flag logic, and branch processing. . Memory load/store operations are performed using AGUs 212, 214. For an embodiment, The integer ALUs 216, 218, 220 are illustrated in the context of the integer operation of the 64-bit metadata operand. In various embodiments, ALUs 216, 218, 220 can be implemented to support multiple data bits including 16, 32, 128, 256, and the like. Similarly, floating point units 222, 224 can be implemented to support operand ranges having bits of various widths. For an embodiment, the floating point units 222, 224 can operate on a 128-bit wide packed data operand in conjunction with SIMD and multimedia instructions.

In one embodiment, the micro-ops schedulers 202, 204, 206 dispatch slave operations before the parental load has finished executing. As the micro-operations are speculatively scheduled and executed in the processor 200, the processor 200 also contains logic to handle memory misses. If a data load is missed in the data cache, it may be that the slave operation has left the scheduler temporarily incorrect in the pipeline trip. A replay organization tracks and re-executes instructions for using incorrect data. Only slave operations will need to be replayed and one of the independents will be allowed to complete. The scheduler and replay of an embodiment of a processor are also designed to capture instructions that provide a base register exchange state verification function.

The name "scratchpad" may refer to an on-board processor storage location that is used as part of the instruction that identifies the operand. In other words, the scratchpad can be those that are available from outside the processor (from a programmer's point of view). However, the register of an embodiment should not be limited in its sense to a particular type of circuit. Rather, the register of an embodiment is storable and provides information and performs the functions described herein. The registers described herein can be dynamically implemented by a physical scratchpad by using any number of different techniques by circuitry implemented within a processor, such as a dedicated physical scratchpad, renamed using a scratchpad. , dedicated and dynamic combination is assigned to the physical register, and so on. In one embodiment, the integer register stores 32-bit integer data. In one embodiment, the scratchpad file also contains eight multimedia SIMD registers for packaging the data. For the discussion below, the scratchpad is considered to be a data buffer designed to hold the package data, for example, 64 bits in a microprocessor enabled by MMX technology from Intel Corporation of Santa Clara, California, USA. Yuan Xun's MMX ^TM scratchpad (also referred to as the 'mm' register in some instances). These MMX registers are available in both integer and floating point formats and can be operated with packaged data elements that accompany SIMD and SSE instructions. Similarly, a 128-bit wide XMM register associated with SSE2, SSE3, SSE4, or later (collectively "SSEx") technology can also be used to maintain such packaged data operands. In an embodiment, in storing the package data and the integer data, the registers do not need to distinguish between the two data types. In one embodiment, integers and floating points are included in the same scratchpad file or in different scratchpad files. Further, in an embodiment, floating point and integer data can be stored in different registers or in the same register.

In the graphic example below, some data operands are illustrated. FIG. 3A illustrates various package data type representations in a multimedia buffer in accordance with an embodiment of the present invention. 3A illustrates a data pattern for a packaged byte 310, a package block 320, and a packaged dword 330 for one of the 128-bit wide operands. The encapsulated byte format 310 of this example is 128 bits long and contains sixteen encapsulated byte data elements. One The tuple is defined here as 8 data bits. The information for each tuple data element is stored in bit 7 to bit 0 for byte 0, bit 15 to bit 8 for byte 1, for byte 2 Bits 23 through 16, and finally bits 120 through 127 for byte 15. Therefore, all available bits in the scratchpad are used. This memory configuration increases the storage performance of the processor. Similarly, by accessing sixteen data elements, an operation can then be performed in parallel on sixteen data elements.

Typically, a data element is an individual piece of data stored in a single scratchpad or memory location with other data elements of the same length. In the package data sequence for SSEx technology, the number of data elements stored in an XMM scratchpad is 128 bits divided by the bit length of another data element. Similarly, in the package data sequence for MMX and SSE techniques, the number of data elements stored in an MMX register is 64 bits divided by the bit length of another data element. Although the data pattern illustrated in FIG. 3A is 128 bits long, embodiments of the present invention may also operate with 64-bit wide, 256-bit wide, 512-bit wide, or other sized operands. The encapsulated block format 320 of this example is 128 bits long and contains eight packed block data elements. Each package block contains sixteen information bits. The packaged double word format 330 of Figure 3A is 128 bits long and contains four encapsulated double word data elements. Each packaged double word data element contains 32 information bits. A package quadword is 128 bits long and contains two encapsulated quadword data elements.

Figure 3B illustrates a data store format in an additional register. Each package The data can contain more than one independent data element. Three package data formats are illustrated; a half package 341, a single package 342, and a dual package 343. One embodiment of the half package 341, the single package 342, and the dual package 343 contains fixed point data elements. For a different embodiment, one or more of the half package 341, the single package 342, and the dual package 343 may contain floating point data elements. A different embodiment of one of the half packages 341 is 128 bits long and contains eight 16 bit data elements. One embodiment of a single package 342 is 128 bits long and contains four 32-bit data elements. One embodiment of dual package 343 is 128 bits long and contains two 64 bit data elements. It should be appreciated that such package data formats can be further extended to other scratchpad lengths, for example, to 96-bit, 160-bit, 192-bit, 224-bit, 256-bit, 512. - bit or more.

3C illustrates various signed and unsigned package data pattern representations in a multimedia buffer in accordance with an embodiment of the present invention. The unsigned packaged byte representation 344 illustrates a memory of an unsigned packaged byte in a SIMD register. The information for each tuple data element is from bit 7 to bit 0 of byte 0, bit 15 to bit 8 of byte 1, bit 23 to bit 16 of byte 2, etc. And the manner of bit 127 to bit 120 of the last byte 15 is stored. Therefore, all available bits in the scratchpad are used. This storage configuration increases the storage performance of the processor. Similarly, by accessing the sixteen data elements, an operation can then be performed in parallel on sixteen data elements. The signed packed byte representation 345 illustrates the storage of a signed packed byte. Note It is intended that the eighth bit of each byte data element is a symbol indicator. The unsigned package block indicates 346 how the block 7 to block 0 are stored in a SIMD register. The signed packed block representation 347 is an unsigned packed block similar to the representation 346 in the scratchpad. Note that the sixteenth bit of each block data element is a symbol indicator. The unsigned encapsulation double-word representation 348 shows how the double-word data elements are stored. The signed packed double block representation 349 is an unsigned packed double block similar to the representation 348 in the scratchpad. Note that the required sign bit is the thirty-second bit of each double-word data element.

3D is a representation of one embodiment of an opcode format 360 having thirty-two or more bits, and one of the scratchpad/memory operand addressing modes, which corresponds to being described in "Intel® 64. And a type of opcode format in the IA-32 Intel Architecture Software Developer's Handbook Volumes 2A and 2B: Instruction Set Reference AZ" on the World Wide Web (www) intelcom/products/processor/manuals/ Provided by Intel Corporation of Santa Clara, California. In one embodiment, the instructions may be encoded using one or more fields 361 and 362. Up to two operand positions per instruction can be identified, including up to two source operand identifiers 364 and 365. For an embodiment, the destination operand identifiers 366 are the same as the source operand identifiers 364, while in other embodiments they are different. For a different embodiment, the destination operand identifiers 366 are the same as the source operand identifiers 365, while in other embodiments they are different. In one embodiment, one of the source operands identified by the source operand identifiers 364 and 365 is overlaid with the result of the instruction. Write, while in other embodiments, the identifier 364 corresponds to a source register element and the identifier 365 corresponds to a destination register element. For an embodiment, operand identifiers 364 and 365 can be used to identify 32-bit or 64-bit sources as well as destination operands.

FIG. 3E is a representation of another different opcode format 370 having 40 or more bits. The opcode format 370 corresponds to the opcode format 360 and includes a select prefix prefix 378. Instructions in accordance with an embodiment may be encoded using one or more of the fields 378, 371, and 372. Up to two operand positions per instruction can be identified using source operand identifiers 374 and 375 and using prefix first byte 378. For an embodiment, the prefix byte 378 can be used to identify a 32-bit or 64-bit source and destination operand. For an embodiment, the destination operand identifiers 376 are the same as the source operand identifiers 374, while in other embodiments they are different. For a different embodiment, the destination operand identifiers 376 are the same as the source operand identifiers 375, while in other embodiments they are different. In one embodiment, an instruction operates on one or more operands identified by operand identifiers 374 and 375 and the one or more operands identified using operand identifiers 374 and 375 utilize the result of the instruction. The writes are overwritten, while in other embodiments, the operands identified by the identifiers 374 and 375 are written to another material element in another register. Opcode formats 360 and 370 allow the scratchpad to the scratchpad, the memory to the scratchpad, the scratchpad by the memory, the scratchpad by the scratchpad, the instant register, the scratchpad to Memory addressing, which uses partial MOD columns 363 and 373 and utilization options The scalar-index-base and shift byte are specified.

Turning next to Figure 3F, in some different embodiments, 64-bit (or 128-bit, or 256-bit, or 512-bit or more) single instruction majority data (SIMD) arithmetic operations may be via one Auxiliary Processor Data Processing (CDP) instructions are performed. The opcode format 380 shows such a CDP instruction with CDP opcode columns 382 and 389. The version of the CDP instruction, for different embodiments, may be encoded using one or more of the fields 383, 384, 387, and 388. Each instruction can have up to three operand locations identified, including up to two source operand identifiers 385 and 390 and a destination operand identifier 386. One embodiment of the secondary processor can operate on 8, 16, 32, and 64 bit values. For an embodiment, an instruction is performed on an integer data element. In some embodiments, an instruction may be conditionally executed using condition bar 381. For some embodiments, the source material size can be encoded using field 383. In some embodiments, zero (Z), negative (N), carry (C), and overflow (V) detections can be done on the SIMD column. For some instructions, the saturated version can be encoded using column 384.

Turning next to Figure 3G, which is a representation of another different opcode format 397 to provide a substrate scratchpad exchange state verification function in accordance with another embodiment, the opcode format 397 corresponding to The version of the opcode format in the Intel® Advanced Vector Extension Stylization Reference is available on Intel Worldwide at www.intelcom/products/processor/manuals/ by Intel Corporation of Santa Clara, California.

An original x86 instruction set provided for having an address byte of various formats included in the additional byte and a 1-byte operation code of the immediate operand, the presence of the first "opcode" The byte was informed. Additionally, there are certain byte values that are reserved as modifiers to the opcode (referred to as prefixes since they must be placed before the instruction). When the original palette of 256 opcoded tuples (including these special word first values) is exhausted, a single byte is dedicated as a escape to the new set of 256 opcodes. As vector instructions (eg, SIMD) are added, more opcodes are required to be generated, and even when expanded by use of word prefixes, "2-byte" opcode mapping is not sufficient. For this purpose, new instructions are added to the additional map, which uses 2 bytes plus a select prefix as an identifier.

Additionally, to facilitate additional registers in the 64-bit mode, an additional prefix can be used (referred to as "REX") at the beginning of the prefix and the opcode (and any necessary to determine the opcode) Between the escaped bytes). In one embodiment, the REX may have four "payload" bits to indicate the use of additional scratchpads in the 64-bit mode. In other embodiments, it may have less than or more than 4 bits. The general format of at least one instruction set (which typically corresponds to format 360 and/or format 370) is typically instantiated in the following manner: [prefixes][rex]escape[escape2]opcode modrm(etc.)

The opcode format 397 corresponds to the opcode format 370 and includes a selector VEX prefix first byte 391 (starting with C4 (hexadecimal) in one embodiment) to replace most other conventional instruction prefix first tuples that are typically used. And escape digital. For example, the following illustrates an embodiment in which two fields are used to encode an instruction when a second escape digit is rendered with the original instruction or when additional bits in the REX column (eg, XB and W columns) will Can be used when it needs to be used. In the exemplified embodiment below, the traditional escape is represented by a new escape value, which is completely compressed as part of the "payload" byte, the traditional prefix being recovered and available for the future. The second escape number is compressed in a "mapping" column with additional mapping or feature space available, and new features are added (eg, increasing the vector length and an additional source register specifier).

Instructions in accordance with an embodiment may be encoded using one or more fields 391 and 392. Each instruction can have up to four operand positions for combination by source operand identifiers 374 and 375 and a selective scalar-index-base (SIB) identifier 393, a selective displacement identifier 394. And a combination of a selective instant byte 395 is identified using column 391. For an embodiment, the VEX prefix first byte 391 can be used to identify 32-bit or 64-bit source and destination operands and/or 128-bit or 256-bit IMD registers or memory operands. For an embodiment, the functionality provided using opcode format 397 may be redundant for opcode format 370, while in other embodiments they are different. Fuck Code formats 370 and 397 allow the scratchpad to the scratchpad, the memory to the scratchpad, the scratchpad by the memory, the scratchpad by the scratchpad, the instant register, the scratchpad to The memory is addressed, and the memory address is specified in part using the MOD column 373 and using a selection formula (SIB) identifier 393, a selective displacement identifier 394, and a selective instant byte 395.

Turning next to Figure 3H, which is a representation of another different opcode format 398 for providing a substrate scratchpad exchange status verification function in accordance with another embodiment. Opcode format 398 corresponds to opcode formats 370 and 397 and includes a selective EVEX prefix first byte 396 (starting with 62 (hexadecimal) in one embodiment) to replace most of the other conventional instruction words that are typically used. The first tuple and the escape digits provide additional functionality. Instructions in accordance with one embodiment may be encoded using one or more of fields 396 and 392. Up to four operand positions per instruction and a mask can be combined by field 396 and source operand identifiers 374 and 375 and with a selective scalar-index-base (SIB) identifier 393, A combination of a selective displacement identifier 394 and a selective instant bit 395 is identified. For an embodiment, the EVEX prefix first byte 396 can be used to identify 32-bit or 64-bit sources as well as destination operands and/or 128-bit, 256-bit or 512-bit IMD registers or memory. Operator. For an embodiment, the functionality provided by opcode format 398 may be redundant to opcode format 370 or 397, while in other embodiments they are different. The opcode format 398 allows the scratchpad to the scratchpad, the memory to the scratchpad, the scratchpad by the memory, the scratchpad by the scratchpad, the instant register, the scratchpad to the memory Addressing, with a mask, the memory The addressing is partially specified using the MOD column 373 and using the selection formula (SIB) identifier 393, a selective displacement identifier 394, and a selective instant byte 395. The general format of at least one instruction set (which typically corresponds to format 360 and/or format 370) is typically instantiated by: evex1 RXBmmmmm WvvvLpp evex4 opcode modrm[sib][disp][imm]

For an embodiment, an instruction encoded in accordance with EVEX format 398 may have additional "pay" bits that may be used to provide a base register exchange status verification function, for example, with additional new features, for example, A user configurable mask register, or an additional operand, or between 128-bit, 256-bit or 512-bit vector registers, or more temporary memory selected from , and so on.

Examples of instructions, some of which can be used to provide a substrate scratchpad exchange state verification function, are illustrated by way of example and are not limited to the following list:

It should be appreciated that the base register exchange state verification instructions, as in the above examples, can be used to provide exception handler inference capabilities when executed, or for example, whether or not a SwapGS instruction needs to be executed without having to resort to an ad hoc method, For example, a complex and vulnerable "magic address check", or setting the GS base register to a negative value in the core and using the user space command to set the FS base register and the GS base register. can. It should also be appreciated that the base register exchange status verification instructions, as in the above examples, can be used to avoid errors in exception handler design, additional time consuming checks, and unnecessary user restrictions.

4A is a block diagram illustrating a sequential pipeline and a scratchpad renaming step, an out-of-order issue/execution pipeline, in accordance with at least one embodiment of the present invention. 4B is a block diagram illustrating a sequential structure core and a scratchpad rename logic, out-of-order issue/execution logic included in a processor in accordance with at least one embodiment of the present invention. The solid line block in Figure 4A illustrates the sequential pipeline, while the dashed line block illustrates the register renaming, out of order release/execution pipeline. Similarly, the solid line block in FIG. 4B illustrates sequential structure logic, while the dashed line block illustrates scratchpad rename logic and out-of-order issue/execution logic.

In FIG. 4A, a processor pipeline 400 includes a capture step 402, a length decoding step 404, a decoding step 406, an assigning step 408, a renaming step 410, and a schedule (also known as a dispatch or Release 412, a scratchpad read/memory read step 414, an execution step 416, a writeback/memory write step 418, an exception processing step 422, and a determination step 424.

In Fig. 4B, an arrow indicates a coupling between two or more units and an arrow direction indicates the direction of the data flow between those units. 4B shows a processor core 490 that includes a front end point unit 430 coupled to an execution engine unit 450, and both of which are coupled to a memory unit 470.

Core 490 can be a simplified instruction set computing (RISC) core, a complex instruction set computing (CISC) core, a very long instruction block (VLIW) core, or a hybrid or different core type. Alternatively, the core 490 can be a special purpose core, such as a network or communication core, a compression engine, a graphics core, or the like.

The pre-endpoint unit 430 includes a branch prediction unit 432 coupled to an instruction cache unit 434 that is coupled to an instruction conversion lookaside buffer (TLB) 436 that is coupled to an instruction fetch unit 438, which is coupled to a decoding unit 440. The decoding unit or decoder may decode the instructions and, as an output, generate one or more micro-ops, microcode entry points, microinstructions, other instructions, or other control signals that are decoded from the original instructions, or otherwise Reflect the original instruction or be exported from the original instruction. The decoder can be implemented using a variety of different mechanisms. Examples of suitable mechanisms include, but are not limited to, look-up tables, hardware implementations, programmable logic arrays (PLAs), microcode read only memory (ROM), and the like. The instruction cache unit 434 is further coupled to a level 2 (L2) cache unit 476 in the memory unit 470. The decoding unit 440 is coupled to one of the execution engine units 450, the rename/allocator unit 452.

Execution engine unit 450 includes the weight coupled to a decommissioning unit 454 A new naming/dispenser unit 452 and a set of one or more scheduler units 456. Scheduler unit 456 represents any number of different schedulers, including reservation stations, central instruction windows, and the like. The scheduler unit 456 is coupled to the physical register file unit 458. Each of the physical scratchpad file units 458 represents one or more physical register files, one of which stores one or more different data types, such as scalar integers, scalar floating points, packed integers , wrap floating point, vector integer, vector floating point, etc., state (eg, an instruction indicator, which is the address of the next instruction to be executed), and so on. The physical scratchpad file unit 458 overlaps with the decommissioning unit 454 to illustrate various ways in which register renaming and out-of-sequence execution can be performed (eg, using a reorder buffer and a delimited register file, using a future File, a history buffer, and a delimited scratchpad file; use a scratchpad map and a shared scratchpad; etc.). Typically, the structure register is external to the processor or visible from a programmer's point of view. The registers are not limited to any conventional, specific type of circuitry. A variety of different types of registers are suitable as long as they are capable of storing and providing information as explained herein. Examples of suitable registers include, but are not limited to, a combination of a dedicated entity register, a physical register that is dynamically allocated using a register renaming, a dedicated, and a dynamically allocated physical register. ,and many more. The decommissioning unit 454 and the physical register file unit 458 are coupled to the execution aggregate 460. Execution aggregation 460 includes a set of one or more execution units 462 and a set of one or more memory access units 464. The execution unit 462 can perform various operations (eg, shifting, addition, subtraction, multiplication) as well as in various types Various operations (for example, scalar floating point, packed integer, encapsulated floating point, vector integer, vector floating point) perform various operations. Although some embodiments may include some execution units that are specific to a particular function or group of functions, other embodiments may include only one execution unit or a plurality of execution units that perform all of the functions. Scheduler unit 456, physical register file unit 458, and execution aggregate 460 are shown as complex numbers, as some embodiments generate individual pipelines for use in certain types of data/operations (eg, each having Their unique scheduler unit, physical register file unit, and/or execution of a scalar integer pipeline, a scalar floating point/packaged integer/packaged floating point/vector integer/vector floating point pipeline, and / or a memory access pipeline, and in the case of a separate memory access pipeline, some embodiments in which only the aggregation of this pipeline with memory access unit 464 is performed are performed). It should also be understood that where separate pipelines are used, one or more of these pipelines may be out of order for release/execution and the rest are sequential.

The set of memory access units 464 is coupled to a memory unit 470 that includes a data TLB coupled to a data cache unit 474 that is coupled to a one-bit 2 (L2) cache unit 476. Unit 472. In one embodiment, the memory access unit 464 can include a load unit, a memory address unit, and a stored data unit, each coupled to a data TLB unit 472 in the memory unit 470. The L2 cache unit 476 is coupled to one or more other levels of cache and eventually to a primary memory.

By example, the sample register is renamed, out of order release/execution core The heart structure may implement pipeline 400 as follows: 1) instruction capture 438 for capture and length decoding steps 402 and 404; 2) decoding unit 440 for decoding step 406; 3) rename/allocator unit 452 for allocation step 408 and renaming step 410; 4) scheduler unit 456 performs scheduling step 412; 5) physical register file unit 458 and memory unit 470 perform register read/memory read step 414; perform aggregation 460 performs step 416; 6) memory unit 470 and physical register file unit 458 performs write back/memory write step 418; 7) various units may involve exception processing step 422; and 8) decommissioning unit 454 and The physical scratchpad file unit 458 performs a determination step 424.

Core 490 can support one or more instruction sets (eg, the x86 instruction set (with some extensions that have been added with newer versions); MIPS instruction set for MIPS technology in Sunnyvale, California; Sunnyvale, California ARM's ARM instruction set (with optional extensions, such as NEON).

It should be appreciated that the core can support multiple threads (performing two or more parallel operations or threads) and can do so in a variety of ways, including time-sharing multi-threading, simultaneous multi-threading (where a single actual core provides a logical core to Each of the threads for the actual core of simultaneous multi-threading, or a combination thereof (eg, time-sharing and decoding and, for example, subsequent multi-threading as in Intel® Hyper-Threading Technology).

Although register renaming is illustrated in the context of out-of-order execution, it should be understood that register renaming can be used in a sequential structure. Although the illustrated processor embodiment also includes a separate instruction and data cache unit 434/474 and a shared L2 cache unit 476, but different embodiments may also have, for example, a single internal cache for both instructions and data, for example, a one-bit 1 (L1) internal cache, or Most of the internal caches are standard. In some embodiments, the system can include an internal cache and a combination of external caches at one of the core and/or outside of the processor. Additionally, all caches may be external to the core and/or processor.

5 is a block diagram of a single core processor and a multi-core processor 500 having an integrated memory controller and graphics, in accordance with an embodiment of the present invention. The solid line block in FIG. 5 illustrates a processor 500 having a single core 502A, a system agent 510, a set of one or more bus controller units 516, and a dashed line of options added to illustrate a different process. The device 500 has a plurality of cores 502A-N, one or more integrated memory controller units 514 in one of the system proxy units 510, and an integrated graphics logic 508.

The memory hierarchy includes one or more cache levels within the core, a set or one or more shared cache units 506, and external memory coupled to the set of integrated memory controller units 514 (not Shown in the graph). The set of shared cache units 506 may include one or more intermediate level caches, such as level 2 (L2), level 3 (L3), level 4 (L4), or other cache level, one. Last level cache (LLC), and/or combinations thereof. Although in one embodiment, a ring substrate interconnect unit 512 interconnects the integrated graphics logic 508, the set of shared cache units 506, and the system proxy unit 510, different embodiments may use any number of conventional techniques to Interconnect these units.

In some embodiments, one or more of the cores 502A-N can be multi-threaded. The system agent 510 includes those components that coordinate and operate the cores 502A-N. The system proxy unit 510 can include, for example, a power control unit (PCU) and a display unit. The PCU can be or include the logic and components needed to adjust cores 502A-N and integrate graphics logic 508. The power state of the display unit is a display used to drive one or more external connections.

The cores 502A-N may be homogeneous or heterogeneous in terms of structure and/or instruction set. For example, some cores 502A-N may be sequential and others may be out of order. As another example, two or more cores 502A-N may be capable of executing the same set of instructions, while others may be capable of executing only a subset of the set of instructions or a different set of instructions.

The processor may be a general purpose processor, e.g., Core ^TM i3, i5, i7,2Duo and ^{Quad, Xeon TM, Itanium TM,} XScale TM or StrongARM ^TM processor, which is provided by the Intel Corporation of Santa Clara, Gary of. Additionally, the processor may be from another company, such as ARM Holdings, MIPS, and the like. For example, the processor can be a special purpose processor, such as a network or communications processor, a compression engine, a graphics processor, a secondary processor, an embedded processor, or the like. The processor can be implemented on one or more wafers. The processor 500 can be a component and/or can be implemented on one or more substrates, for example, using any processing technique, such as BiCMOS, CMOS, or NMOS.

6-8 are examples of systems suitable for use with processor 500, and FIG. 9 is an example of a system chip (SoC) that may contain one or more cores 502. For laptops, desktops, handheld PCs, personal digital assistants, engineering workstations, servers, networking devices, network hubs, switches, embedded processors, digital, signal processor (DSP), graphics devices Other system designs and configurations of conventional techniques for video game devices, set-top boxes, microcontrollers, cell phones, portable media players, handheld devices, and various other electronic devices are also suitable. In general, a wide variety of systems or electronic devices, as disclosed herein that may include a processor and/or other execution logic, are also generally suitable.

Referring next to Figure 6, a block diagram of a system 600 in accordance with an embodiment of the present invention is shown. The system 600 can include one or more processors 610, 615 that are coupled to a graphics memory controller hub (GMCH) 620. The alternative nature of the additional processor 615 is indicated in Figure 6 by a dashed line.

Each processor 610, 615 can be some version of processor 500. However, it should be noted that it is not possible to have integrated graphics logic and integrated memory control units present in the processors 610, 615. The GMCH 620 illustrated in Figure 6 can be coupled to a memory 640, which can be, for example, a dynamic random access memory (DRAM). For at least one embodiment, the DRAM can be associated with a non-electrical cache.

The GMCH 620 can be a chipset, or a portion of a wafer set. The GMCH 620 can communicate with the processors 610, 615 and control the interaction between the processors 610, 615 and the memory 640. The GMCH 620 can also function as the processors 610, 615 and other elements in the system 600. An accelerated bus interface between the pieces. For at least one embodiment, the GMCH 620 communicates with the processors 610, 615 via a multi-drop branch bus (e.g., a front side bus (FSB) 695).

Still further, the GMCH 620 is coupled to a display 645 (eg, a flat panel display). The GMCH 620 can include an integrated graphics accelerator. The GMCH 620 is further coupled to an input/output (I/O) controller hub (ICH) 650 that can be used to couple various peripheral devices to the system 600. For example, shown in the embodiment of FIG. 6, an external graphics device 660, which may be coupled to another peripheral device 670, to a discrete graphics device of the ICH 650.

Additionally, additional or different processors may also be present in system 600. For example, the additional processor 615 can include additional processors identical to the processor 610, additional processors that are heterogeneous or asymmetric to the processor 610, acceleration devices (eg, graphics acceleration devices or digital signal processing (DSP) Unit), field programmable gate array, or any other processor. There are many differences between actual resources 610, 615 in terms of spectrum including the structure, microstructure, heat, power consumption characteristics, and the like. These differences may effectively present themselves as asymmetry and non-uniformity among the processors 610, 615. For at least one embodiment, various processors 610, 615 may be present in the same wafer package.

Referring next to Figure 7, shown is a block diagram of a second system 700 in accordance with an embodiment of the present invention. As shown in FIG. 7, multiprocessor system 700 is a point-to-point interconnect system and includes a first processor 770 and a second processor 780 coupled via a point-to-point interconnect 750. deal with Each of the devices 770 and 780 can be some version of the processor 500 as one or more processors 610, 615.

Although only two processors 770, 780 are shown, it should be understood that the scope of the invention is not so limited. In other embodiments, one or more additional processors may reside in a given processor.

Processors 770 and 780 are shown to include integrated memory controller units 772 and 782, respectively. Processor 770 also includes components of its bus controller unit point-to-point (P-P) interfaces 776 and 778; likewise, second processor 780 includes P-P interfaces 786 and 788. Processors 770, 780 can use P-P interface circuits 778, 788 to exchange information via a point-to-point (P-P) interface 750. As shown in FIG. 7, IMCs 772 and 782 couple the processors to separate memories, namely a memory 732 and a memory 734, which may be localized to the respective processors. Part of the memory.

Processors 770, 780 can each exchange information with a chipset 790 via respective P-P interfaces 752, 754 using point-to-point interface circuits 776, 794, 786, 798. Wafer set 790 can also exchange information with a high performance graphics circuit 738 via a high performance graphics interface 739.

A shared cache (not shown in the graphics) can be included in a processor or outside of the two processors, still connected to the processor via a PP interconnect, so that if a processor is placed low In the power mode, the localized cache information of any or two processors can be stored in the shared cache.

Wafer set 790 can be coupled to a first confluence via an interface 796 Row 716. In an embodiment, the first bus bar 716 can be a peripheral component interconnect (PCI) bus bar, or a bus bar, for example, a PCI express bus or another third-generation I/O interconnect bus. However, the scope of the present invention is not limited thereto.

As shown in FIG. 7, various I/O devices 714 can be coupled to first bus bar 716, and a bus bar bridge 718 that couples first bus bar 716 to a second bus bar 720. In an embodiment, the second bus 720 can be a low pin count (LPC) bus. In one embodiment, various devices can be coupled to the second bus 720, for example, including a keyboard and/or mouse 722, a communication device 727, and a storage unit 728, such as a disc drive or other mass storage device. A device, which may include instructions/digitals and data 730. Further, an audio I/O 724 can be coupled to the second bus 720. It should be noted that other configurations are also possible. For example, instead of the point-to-point structure of Figure 7, a system can implement a multi-drop branch bus or other such structure.

Referring next to Figure 8, a block diagram of a third system 800 in accordance with an embodiment of the present invention is shown. Similar elements in Figures 7 and 8 bear the same reference numerals, and certain aspects of Figure 7 are omitted from Figure 8 to avoid confusing the other points of Figure 8.

FIG. 8 illustrates processors 870, 880, which may include integrated memory and I/O control logic ("CL") 872 and 882, respectively. For at least one embodiment, CL 872, 882 can include an integrated memory controller unit, such as described above in connection with Figures 5 and 7. Also. CL 872, 882 can also contain I/O control logic. Figure 8 illustrates more than just being coupled to Memory 832, 834 of CL 872, 882, but also I/O device 814, also coupled to control logic 872, 882. Conventional I/O device 815 is coupled to chip set 890.

Referring next to Figure 9, shown is a block diagram of a SoC 900 in accordance with an embodiment of the present invention. Elements similar to those in Figure 5 have the same reference numbers. At the same time, the dotted squares are also a selective feature on more advanced SoCs. In Figure 9, an interconnect unit 902 is coupled to: an application processor 910 that includes a set of one or more cores 502A-N and a shared cache unit 506; a system proxy unit 510; a bus control Unit 516; an integrated memory controller unit 514; a set of one or more media processors 920, which may include integrated graphics logic 508, an image processor 924 for providing static and/or video camera functions, An audio processor 926 for providing hardware audio acceleration, and a video processor 928 for providing video encoding/decoding acceleration; a static random access memory (SRAM) unit 930; a direct memory access ( DMA) unit 932; and display unit 940 for coupling to one or more external displays.

Figure 10 illustrates a processor including a central processing unit (CPU) and a graphics processing unit (GPU) that can perform at least one instruction in accordance with an embodiment. In one embodiment, an instruction to operate in accordance with at least one embodiment may be performed using a CPU. In another embodiment, the instructions can be made using a GPU. In still another embodiment, the instructions can be made via a combination of operations performed using the GPU and the CPU. For example, in one embodiment, an instruction may be received in accordance with an embodiment and And is decoded for execution on the GPU. However, one or more operations within the decoded instruction may be performed using a CPU and the results are returned to the GPU for the final decommissioning of the instruction. Conversely, in some embodiments, the CPU can function as a primary processor and the GPU acts like a secondary processor.

In some embodiments, instructions that benefit from a highly parallel, capacity processor can be utilized with the GPU, while instructions that benefit from processor performance (which benefits from the deep pipeline structure) can be utilized with the CPU. For example, graphics, scientific applications, financial applications, and other parallel workloads can benefit from the performance of the GPU and are therefore executed, while more sequential applications, such as operating system cores or application numbers, may be better suited for the CPU.

In FIG. 10, the processor 1000 includes a CPU 1005, a GPU 1010, an image processor 1015, a video processor 1020, a USB controller 1025, a UART controller 1030, an SPI/SDIO controller 1035, a display device 1040, and high definition. Multimedia Interface (HDMI) Controller 1045, MIPI Controller 1050, Flash Memory Controller 1055, Dual Data Rate (DDR) Controller 1060, Security Engine 1065, and I ² S/I ² C (Integrated Inter-Wound Sound / Between the integrated circuits) interface 1070. Other logic and circuitry can be included in the processor of Figure 10, including more CPUs or GPUs and other peripheral interface controllers.

One or more arguments of at least one embodiment may be implemented by representing a representation of various logic within the processor by being stored on a machine readable medium, which when read by a machine This causes the machine to be logical to perform the techniques described herein. Such representations, such as "IP cores", can be stored on a physical machine readable medium ("cartridge") and supplied to various customers or production facilities to be loaded into the actual constituent logic or processor. On the manufacturing machine. For example, IP cores, such as the Cortex ^TM family of processors developed by ARM Holdings, and the Loongson IP core developed by the Institute of Computer Technology (ICT) of the Chinese Academy of Sciences, can be licensed or sold to various customers or licensees, for example, Texas Instruments , Qualcomm, Apple, or Samsung, and are implemented in processors that utilize these customers or licensees to be generated.

11 shows a block diagram of an IP core development in accordance with an embodiment. The storage 1130 includes a simulation software 1120 and/or a hardware or software module 1110. In one embodiment, data representative of the IP core design may be provided to the storage 1130 via a memory 1140 (eg, a hard drive), a wired connection (eg, the Internet) 1150, or a wireless connection 1160. The IP core information generated using the simulation tools and modules can then be sent to a manufacturing device that can be manufactured by a third group for at least one instruction in accordance with at least one embodiment.

In some embodiments, one or more instructions may correspond to a first version or structure (eg, x86) and be transformed or simulated on a different type or structure of one of the processors (eg, ARM). An instruction, according to an embodiment, can therefore be implemented on any processor or processor type, including ARM, x86, MIPS, a GPU, or other processor type or architecture.

Figure 12 illustrates how the instructions of the first version according to one embodiment are A different type of processor is used to simulate. In FIG. 12, the program 1205 contains instructions that can perform the same or the same function as an instruction, according to an embodiment. However, the instructions of program 1205 may be different from or incompatible with a type and/or format of processor 1215, which means that the pattern of patterns in program 1205 may not be able to be originally executed by processor 1215. However, with the aid of the emulation logic 1210, the instructions of the program 1205 are translated into instructions that were originally executable by the processor 1215. In one embodiment, the analog logic is implemented in a hardware manner. In another embodiment, the analog logic is implemented in a materialized machine readable medium that includes software to translate the type of instructions into the original usable processor 1215 in the program 1205. The type of execution. In other embodiments, the analog logic is a combination of fixed function or programmable hardware and a program stored on a materialized machine readable medium. In one embodiment, the processor contains analog logic, while in other embodiments, the analog logic is external to the processor and is provided by a third community. In one embodiment, the processor is capable of loading the analog logic, the analog logic being implemented in an entity comprising software that is executed by or associated with the processor or associated with the processor. The machine can be read in the media.

13 is a block diagram of the use of a software command converter for converting a binary instruction in a source instruction set to a binary instruction in a target instruction set in accordance with an embodiment of the present invention. In the illustrated embodiment, the command converter is a software command converter, although the command converter can be additionally implemented in software, firmware, hardware, or various combinations thereof. Figure 13 A program in a higher level language 1302 is shown that can be compiled using an x86 compiler 1304 to produce an x86 binary number 1306 that can be originally executed using a processor having at least one x86 instruction set core 1316 . A processor having at least one x86 instruction set core 1316 represents any processor that can perform the same function as an Intel processor having one of at least one x86 instruction set core, the processor being implemented by compatibility or otherwise in a different manner Handling (1) a major portion of the Intel x86 instruction set core instruction set or (2) application or other software object code version to target on an Intel processor with at least one x86 instruction set core to achieve The same result is the real value of an Intel processor with at least one of the x86 instruction set cores. The x86 compiler 1304 represents a compiler operative to generate x86 binary digits 1306 (eg, object code), possibly with or without additional link processing, executed with at least one x86 instruction set core On the processor of 1316. Similarly, FIG. 13 shows a program in higher order language 1302 that can be compiled using an alternate instruction set compiler 1308 to generate an alternate instruction set binary digit 1310 that can be utilized originally without at least one x86 instruction set core 1314 One of the processors is executed (eg, a core processor that implements the MIPS instruction set for Sunnyvale's MIPS technology in California and/or executes the ARM instruction set of ARM Holdings of Sunnyvale, Calif.). The command converter 1312 is used to convert the x86 binary 1306 into a digital, which can be originally executed using the processor that does not have an x86 instruction set core 1314. This converted digital is very unlikely to be the same as the alternative instruction set. Digit 1310, because an instruction converter that can do this may not be easy to achieve; however, the converted digital will achieve general operation and consist of different instruction sets. Thus, the command converter 1312 represents software, firmware, hardware, or a combination thereof that allows for an processor or other electronic device that does not have an x86 instruction set processor or core via analog, analog, or any other processing program. The device executes the x86 binary 1306.

In some embodiments, a high-level language 1306 can support the use of essential functions, which is a function known to the x86 compiler 1304, which maps directly to a sequence of one or more combined language instructions to pass x86 binary digits 1306. One of the operation codes provides a base register exchange state verification function. It should be appreciated that in some embodiments, an alternate instruction set compiler 1308 and/or instruction converter 1312 can recognize one or more essential functions and/or one or more combined language instructions to provide a base register exchange, respectively. State verification function, and may generate converted digital numbers that will achieve general operations, and/or allow for a processor or other electronic device that does not have an x86 instruction set processor core via analog, emulation, or any other processing program The x86 binary 1306 is executed to provide a base register exchange status verification function, respectively.

Figure 14 illustrates a different embodiment of a system 1400 for converting instructions that provide a base register exchange status verification function. System 1400 is an example of mapping non-original x86 fabric states to any of memory locations in memory 1410 or to processor resources in a raw processor 1414. The original processor 1414 includes a scratchpad file 1416 that contains the structure of the general scratchpad of the original processor structure. Any number of scratchpads can be raised for. In one embodiment of system 1400, all non-original x86 structural states are mapped to memory 1410. For example, the descriptor table 1432 (which may include a wide area descriptor table, a local area descriptor table, and an interrupt descriptor table), a page table 1430 (which stores virtual pairs of actual address translations), and a task status area. Segment 1428, general register 1420 (not shown in the graphics), sector register 1426, control register 1424, and other registers 1422 may represent non-original configuration states.

In order to access any non-original state of construction, a memory access can be made. For example, if a non-original instruction has one of the general registers as an operand, the translator or the converted original instruction performs a memory access mapped to the memory location of the general scratchpad to save Take or update the general scratchpad. The scratchpad in the scratchpad file 1416 can be used by the translator or converter 1412 as a temporary scratchpad to preserve intermediate results, or alternatively, a set of original scratchpads in the scratchpad file 1416 can be directly The ground is mapped to a general scratchpad of the non-original state of the state or for use in other local translator/converter states. The general register 1420 can include a general register of x86 64-bit integers (eg, RAX, RBX, etc.), as well as additional integer general registers defined with the first byte of the REX word. The vector/FP register 1418 may include an 80-bit floating point register, a stream single instruction, a majority data (SIMD) extension (SSE) register, and an additional SSE defined with the first bit of the REX word. Register. Segment register 1426 may include storage locations corresponding to six 16-bit x86 sector registers, which may also include the FS and GS sector registers shown and optionally mapped to memory 1410 And/or processing

It is supported in the device 1414. Control register 1424 can include storage locations corresponding to various control registers defined in the non-original x86 processor architecture. For example, the control register, for example, includes one of LMA (Long Mode Actuation), LME (Long Mode Enable), Extended Feature Enablement Register (EFER), and includes PG ( Paged ) and PE (protected mode enabled) bit CR0, and LDTR (local descriptor table register) and GDTR (wide area descriptor table register) and CR3 register (its stored page table 1430) The base address is shown. Other control registers (e.g., CR1, CR2, CR4, etc.) may likewise be included in control register 1424, which may also optionally include a control register to store a base register swap state. A column (eg, a GS.base exchange state and/or a FS.base exchange state) is shown in Figure 16B. The other registers 1422 contain any remaining x86 fabric registers. For example, the EFLAGS register (which stores conditional digital information and optionally stores a base register exchange status bar, for example, as shown in Figure 16B), an instruction indicator (IP) register (its storage will be The address of the executed instruction), and the module specific register (MSR) can be included in other registers 1422. An embodiment of the other register 1422 includes a Kernel GSBase MSR, a GS.base MSR, an FS.base MSR, and optionally an MSR to store a base register exchange status bar (eg, a GS.base) The exchange state and/or a FS.base exchange state) is as shown in Figure 16B.

Figure 15A illustrates a flow diagram of an embodiment of a handler 1501 for providing a substrate scratchpad exchange status verification function. The processing program 1501 and other processing programs disclosed herein are performed using a processing block, which can be The digital is operated in a dedicated hardware or software or firmware that includes a general purpose machine or a special purpose machine or a combination of the two. In one embodiment, the SwapGS instruction to exchange one of the GS base registers is decoded in processing block 1518. Next, in process block 1520, an exchange is performed that exchanges a GS base register value and a core GS base register value in response to the decoded SwapGS instruction. In processing block 1522, it is determined whether the exchange of the GS base register value and the core GS base register value is successfully completed. If so, then in process block 1524, in response to the determination that the exchange of the GS base register value and the core GS base register value is successfully completed, an exchange GS state is reversed. Otherwise, the processing will end with a response to the determination that the exchange of the GS base register value and the core GS base register value is successfully completed.

Figure 15B illustrates a flow diagram of a different embodiment of a process 1502 for providing a substrate scratchpad exchange status verification function on an exception entry. In process block 1510, the application instructions are processed. In process block 1512, a system call or interrupt or exception occurs, requiring entry into a system handler handler. In process block 1514, the status of an exchange GS status bar is determined. Next, in process block 1516, a determination is made as to whether a SwapGS instruction is required, for example, based on the state of the exchange GS status bar. If a SwapGS instruction is required, then in a processing block 1518, a SwapGS instruction is decoded. Next in process block 1520, an exchange is performed that exchanges a GS base in response to the decoded SwapGS command. The scratchpad value and the one-core GS base register value. In processing block 1522, it is determined whether the exchange of the GS base register value and the core GS base register value is successfully completed. If so, in processing block 1524, in response to the determination that the exchange of the GS base register value and the core GS base register value is successfully completed, an exchange GS state is reversed, or if Handler block 1516 originally determines that a SwapGS instruction is not required, then the process proceeds to processing block 1526 where the system or system handler instruction is processed. Moreover, in some embodiments, in response to the determination that the exchange of the GS base register value and the core GS base register value in the processing block 1522 is not successfully completed, the process may begin with processing in reverse. In block 1514. Additionally, in some embodiments, an interrupt or error handling routine may be requested in response to the GS base register value and the exchange of the core GS base register value being not successfully completed. Mediation.

Figure 15C illustrates a flow diagram of another different embodiment of a handler 1503 for providing a base register exchange status verification function on an exceptional exit. Beginning in process block 1526, the system or system handler instructions are processed. In process block 1530, portions of an application state are restored for return to an application handler. In processing block 1534, the current state of the swap GS status bar is determined and processing proceeds to processing block 1536 where a determination is made as to whether a SwapGS instruction is required, for example, based on the status of the exchange GS status bar. If so, a SwapGS instruction is decoded in handler block 1538. Processing program In block 1540, an exchange is performed that exchanges a GS base register value and a core GS base register value in response to the decoded SwapGS instruction. In processing block 1542, it is determined whether the exchange of the GS base register value and the core GS base register value is successfully completed. If so, then in process block 1544, an exchange GS state is reversed in response to a determination that the exchange of the GS base register value and the core GS base register value is successfully completed. Moreover, in some embodiments, processing may begin in process block 1540 repeatedly until the exchange of the GS base register value and the core GS base register value is successfully completed. Additionally, in some embodiments, the exchange in response to the GS base register value and the core GS base register value is not a successful completion of the determination, an interrupt or error handling routine can be requested to be mediated . After processing block 1544, or if determined in handler block 1536, no SwapGS instruction is required, then a self-system call or interrupt or exception is returned to process block 1548. Next, in process block 1550, the processing of the application instructions is resumed.

It should be appreciated that while the above examples show that an exchange GS state is reversed, handlers 1501-1503 can also be applied to other segment substrates (eg, FS substrates), segment substrate combinations (eg, FS substrate and GS substrate). ) or other segment attributes (for example, one of the CS attributes for MSR). It should also be appreciated that a descriptor base and/or attribute register swap or modify state can also be effectively changed and not reversed to indicate the exchange or modification state of the descriptor base and/or attribute register, and Without departing from the principles of this disclosure or The scope of the patent application scope is attached.

Figure 16A illustrates elements of a processor microstructure 1601 embodiment that executes instructions to provide a substrate scratchpad exchange state verification function. 16A is a block diagram illustrating a sequential pipeline processor core, a register renaming step, and an out-of-order issue/execution pipeline, in accordance with at least one embodiment of the present invention. The solid line block in Figure 16A illustrates the sequential pipeline, while the dashed line block illustrates the register renaming, out of order release/execution pipeline.

The processor microstructure 1601 includes a decoding unit 1640 to decode at least one SwapGS instruction, an execution engine unit 1650, and a memory unit 1670. The decoding unit 1640 is coupled to one of the execution engine units 1650, the rename/allocator unit 1652. The execution engine unit 1650 includes a rename/dispenser unit 1652 coupled to a decommissioning unit 1654 and a set of one or more scheduler units 1656. The scheduler unit 1656 represents any number of different schedulers, including reservation stations, central command windows, and the like. The scheduler unit 1656 can be coupled to a physical register file that includes a vector entity register 1684, a mask entity register 1682, a floating point (FP) entity register 1680, and an EFLAGS entity register 1620. An instruction indicator (IP) entity register 1621, an MRS entity register 1622, a section entity register 1626, a control (CTL) entity register 1624, and an integer entity register 1686. Each physical register file represents one or more physical register files, one of which stores one or more different data types, such as scalar integers, scalar floating points, packed integers, encapsulated floating points, Vector integer, vector floating point, etc., state (eg, an instruction indicator indicating that the next finger will be executed The address of the order), control, descriptor table register, and so on.

Some embodiments of execution engine unit 1650 include a stored data buffer 1699 in which data elements from a scalar register or from a descriptor register can be written for use in a storage operation, or from a SIMD The vector register can be written to a plurality of individual component storage locations of the memory data buffer 1699 at the same time for operation of a data element for a decentralized (or a stream of storage under the mask) operation. For example, use a single micro-operation). It will be appreciated that the data elements stored in these respective storage locations of the storage data buffer 1699 can then be transferred to accommodate newer load operations without having to access external memory. Address generation logic 1694 generates a valid address 1606 from at least one base address 1604, an index 1605, and a offset 1603 (eg, using integer entity register 1686 or provided as instant data). The memory is allocated in the memory data buffer 1699 to retain the data elements corresponding to the generated valid address 1606 for storage by the memory access unit 1664 to the corresponding memory location. The data element generated corresponding to the valid address 1606 is copied into the stored data buffer 1699. A memory access unit 1664 is operatively coupled to the address generation logic 1694 for accessing a memory location via the memory unit 1670 (in the case of a scatter instruction for having an unmasked value) One of the corresponding mask 1608 elements) stores a data element 1609 corresponding to the memory location of a valid address 1606 generated by the address generation logic 1694 in response to a store command or in response to a scatter command. In one embodiment, if their valid address 1606 corresponds to the valid address of the newer load instruction, it is stored. The data elements in the memory data buffer 1699 can be accessed to satisfy newer load instructions outside of the sequence of instruction instructions.

In some embodiments, memory access unit 1664 is operatively coupled to one of TLV 1672 of memory unit 1670 to convert the virtual address to the actual address using page table information (eg, page table 1430). In one embodiment of processor microstructure 1601, the x86 fabric state is supported by hardware using processor microstructure 1601. For example, the descriptor table 1432 (which may include a wide area descriptor table, a local area descriptor table, and an interrupt descriptor table) is supported by the scratchpad, page table 1430 (which stores the virtual pair actual address) The conversion is supported by a hardware pager stepper, the general register 1420 is supported by the integer entity register 1686, the section register 1426 is supported by the section entity register 1626, and the register is controlled. 1424 is supported by control entity register 1624, and other registers 1422 are supported by EFLAGS 1620, IP 1621, MSR entity register 1622, and the like.

It should be appreciated that the base register swap state verification, as in the embodiments described herein, can be used to provide exception handler capabilities, for example, to infer at runtime whether a SwapGS instruction needs to be executed without having to resort to an ad hoc Method, for example, a complex and vulnerable "magic address check", or setting the GS base register in the core to a negative value and used to set the user space of the FS base register and/or GS base register The instruction is disabled.

Figure 16B illustrates the detailed elements of an embodiment of a processor microstructure 1602 that performs instructions for providing a substrate scratchpad exchange state verification function. Processor microstructure 1602 includes integer entity register 1686, address generation logic 1694, stored data buffer 1699 and memory access unit 1664, etc., perform functions similar to those described above with respect to FIG. 16A. The embodiment of processor microstructure 1602 can also have an MSR physical register 1622 to store a first base address field corresponding to one of the segments GS 1628 for use in a first execution context (eg, an application) (eg, GS.base1630). A second MSR physical register 1622 can store a second base address field (e.g., Kernel GSBase 1632) corresponding to one of the segments GS 1628 for use in a second execution context (e.g., a system or interrupt handler). In one embodiment, a third register stores one of the base register swap status fields 1615 of the extent GS 1628 corresponding to the first and second execution contexts. In various embodiments, the base register swap status column 1615 corresponds to either section FS1627, or both GS 1628 and FS 1627. In some different embodiments, the third register storing the scratchpad swap status column 1615 can be one of the control entity registers 1624 (eg, CR 1625) or the MSR physical register 1622 (eg, MSR 1635). One or a flag register (eg, EFLAGS 1620 or a Rex extension flag RFLAGS register) or some other scratchpad, such as an extended feature enable register (EFER).

The embodiment of MSR entity register 1622 may also store a 64-bit base address field corresponding to one of the segments (eg, FS1627 or GS1628) for at least one execution mode (eg, long mode) (eg, FS) .base1634 or GS.base1630). An embodiment of the MSR physical register 1622 can further store a segment corresponding to at least one other execution mode (eg, a compatibility mode) (eg, FS1627 or GS1628) A 32-bit base address field (eg, substrate 1637 or substrate 1638). As shown in section 1610, one of the at least one execution mode (eg, a compatibility mode) can use a 16-bit section selector (eg, FS 1627 or GS 1628) and include a base address field. A segment descriptor corresponding to one of (for example, substrate 1637 or substrate 1638) and one of the attribute columns (not shown in the graphic). However, one of the at least one additional execution mode (eg, a 64-bit mode) may use a 16-bit segment selector (eg, FS1627 or GS1628) and only include a 64-bit base address bar. A segment descriptor corresponding to one of (for example, FS.base1634 or GS.base1630).

A decoding unit (e.g., 1640) can decode an exchange instruction (e.g., SwapGS), and an execution unit (e.g., 1650) performs one of the exchange of the first MSR value and the second MSR value in response to the decoded exchange instruction. . If the exchange of the first MSR value and the second MSR value is successfully completed, a value of one of the base register exchange status fields 1625 is changed in response to one of the exchanges being successfully completed. It should be appreciated that by providing a value in the base register exchange status column 1625 to indicate whether the first MSR value and the second MSR value are exchanged, the system or interrupt handler context can infer a SwapGS instruction at runtime. Whether it needs to be executed.

It should be appreciated that the base register exchange status verification instructions can be used to avoid complications, additional time consuming inspections, and unnecessary user restrictions in the exception handler design. Other embodiments are also possible and can be considered carefully.

The embodiments of the mechanisms disclosed herein may be practiced in the form of hardware, software, firmware, or a combination of such methods. Embodiments of the invention may be practiced, a computer program executed on a programmable system including at least one processor, a storage system (including electrical and non-electrical memory and/or storage elements), at least one input device, and at least one output device Code.

The code can be applied to input commands to perform the functions described herein and to generate output information. The output information can be applied to one or more output devices in a conventional form. For the purposes of this application, a processing system includes any system having a processor, such as a digital signal processor (DSP), a microcontroller, an application specific integrated circuit (ASIC), or a microprocessor.

The code can be implemented in a higher level program or object oriented programming language to communicate with a processing system. The code can also be implemented in a combination of language or machine language, if desired. In fact, the institutions described herein are not limited to any particular programming language. In any case, the language can be a compiled or translated language.

One or more arguments of at least one embodiment can be implemented by a representation instruction stored on a machine readable medium, the representation instructions representing various logic within the processor, the instructions being utilized by a machine When read, it will cause the machine to make logic to perform the techniques described herein. Such representations, such as "IP core", can be stored on a physical machine readable medium and supplied to various customers or manufacturing facilities for loading into a manufacturing machine that actually constitutes the logic or processor. in.

Such machine readable storage media may include, but are not necessarily limited to, a non-transitory materialized configuration of an article manufactured or formed by a machine or device, including a storage medium, eg, a hard disk, any other type Disc Films, including floppy disks, optical disks, compact disc-read only memory (CD-ROM), rewritable compact discs (CD-RW), and magnetic optical discs, semiconductor devices, for example, read only Memory (ROM), random access memory (RAM), for example, dynamic random access memory (DRAM), static random access memory (SRAM), erasable programmable read only memory (EPROM), fast Flash memory, electrical eliminates programmable read-only memory (EEPROM), magnetic or optical cards, or any other type of media used to store electronic commands.

Accordingly, embodiments of the present invention also include non-transitory materialized machine readable media containing instructions or containing design material, such as hardware description language (HDL), which defines the structures, circuits, devices, and Processor and / or system features. These embodiments may also be referred to as program products.

In some cases, an instruction converter can be used to convert an instruction from a source instruction set to a target instruction set. For example, the instruction converter can convert (eg, use static binary conversion, dynamic binary conversion including dynamic compilation), morph, simulate, or otherwise convert an instruction to one or more other instruction. The command converter can be implemented in software, hardware, firmware, or a combination thereof. The instruction converter can be on the processor, external to the processor, or partially on the processor, and partially external to the processor.

Thus, techniques for making one or more instructions in accordance with at least one embodiment are disclosed. While certain embodiments of the invention have been shown and described in the drawings, it should be understood that The particular construction and configuration shown and illustrated, as various other modifications are possible when studying this disclosure. For example, in this technical area, its growth is rapid and further advancement is not easily foreseen, and the disclosed embodiments can be easily modified by configuration and detail without departing from the disclosure. The principle or scope of the patent application scope.

200‧‧‧ processor

201‧‧‧ Front end point

202‧‧‧Quick Scheduler

203‧‧‧ Out-of-order execution engine

204‧‧‧Slow/general floating point scheduler

206‧‧‧Simple floating point scheduler

208‧‧‧Integer Scratchpad File/Bypass Network

210‧‧‧Floating point register file/bypass network

211‧‧‧Executive block

212‧‧‧ Address Generation Unit (AGU)

214‧‧‧ Address Generation Unit (AGU)

216‧‧‧fast ALU

218‧‧‧fast ALU

220‧‧‧Slow ALU

222‧‧‧Floating ALU

224‧‧‧Floating point mobile unit

226‧‧‧Instruction prefetcher

228‧‧‧ instruction decoder

230‧‧‧ Tracking cache

232‧‧‧Microcode ROM

234‧‧‧Micromanipulation queue

Claims (27)

A processor, comprising: a decoding unit for decoding an instruction having a column for identifying a universal source register; an execution unit for executing the decoded instruction to copy the identified universal source temporary storage The contents of the device to the session register. The processor of claim 1, wherein the instruction is a WRFSBASE instruction. The processor of any one of claims 1 and 2, wherein the sector register is an FS base. The processor of any one of claims 1 to 3, wherein the recognized universal source register is a 64-bit general purpose register. The processor of any one of claims 1 to 3, wherein the recognized universal source register is a 32-bit general purpose register, and the identified content of the universal source register is to be copied Go to the lower 32 bits of the session register. The processor of claim 1, wherein the instruction is a WRGSBASE instruction. The processor of any one of claims 1 and 6, wherein the sector register is a GS substrate. The processor of any one of claims 1, 6, and 7, wherein the recognized universal source register is a 64-bit general purpose register. The processor of any one of claims 1, 6, and 7, wherein the recognized universal source register is a 32-bit general purpose register and is recognized The other content of the universal source register is to be copied into the lower 32 bits of the sector register. A method comprising the steps of: decoding an instruction having a field to identify a universal source register; executing the decoded instruction to copy the identified content of the universal source register to a segment register. The method of claim 10, wherein the instruction is a WRFSBASE instruction. The method of any one of clauses 10 and 11, wherein the sector register is an FS substrate. The method of any one of clauses 10 to 12, wherein the identified universal source register is a 64-bit general purpose register. The method of any one of clauses 10 to 12, wherein the identified universal source register is a 32-bit general purpose register, and the identified content of the universal source register is to be copied to Among the lower 32 bits of the sector register. The method of claim 10, wherein the instruction is a WRGSBASE instruction. The method of any of clauses 10 and 15, wherein the sector register is a GS substrate. The method of any one of claims 10, 15 and 16, wherein the identified universal source register is a 64-bit general purpose register. The method of any one of claims 10, 15 and 16, wherein the identified universal source register is a 32-bit general purpose register and The identified content of the universal source register is to be copied into the lower 32 bits of the sector register. A machine readable medium for storing an executable system page containing a first executable instruction that, if executed by a machine, causes the machine to perform the following actions: a decoding unit for An instruction having a field to identify a universal source register is decoded; an execution unit to execute the decoded instruction to copy the identified content of the universal source register to the segment register. The machine readable medium as claimed in claim 19, wherein the instruction is a WRFSBASE instruction. The machine readable medium of any one of claims 19 and 20, wherein the sector register is an FS substrate. The machine readable medium of any one of claims 19 to 21, wherein the recognized universal source register is a 64-bit general purpose register. The machine readable medium of any one of claims 19 to 21, wherein the recognized universal source register is a 32-bit general purpose register, and the identified content of the universal source register is To be copied into the lower 32 bits of the session register. The machine readable medium as claimed in claim 19, wherein the instruction is a WRGSBASE instruction. The machine readable medium of any one of claims 19 and 24, wherein the sector register is a GS substrate. Machine readable medium as claimed in any of claims 19, 24 and 25 The common source register is the 64-bit universal register. The machine readable medium of any one of claims 19, 24, and 25, wherein the recognized universal source register is a 32-bit general purpose register, and the identified universal source register is The content is to be copied into the lower 32 bits of the session register.